You are on page 1of 106

Cop_1083_7mm 10-05-2004 14:29 Pagina 1

ISSN 1020-4555
Tropical crop–livestock systems in conservation agriculture
The Brazilian experience 5 Integrated Crop Management Vol. 5–2007

Tropical crop–livestock systems in conservation agriculture – The Brazilian experience
This publication describes how pasture, fodder and livestock
production have been integrated into conservation agriculture
systems in Brazil’s tropical zones. Vast areas of forest have been
cleared in the tropical areas of Brazil for establishment of pastures
that become unproductive once the native fertility of the soil is
exhausted; this leads to yet more forest clearing.

Integrated crop–livestock zero tillage systems allow for the
sustainable production of high-yielding pasture without further
deforestation; in this system, grazing livestock convert both pastures
Tropical crop–livestock systems
and crop residues into cash. The ability of pasture to build up the
fertility and biological activity of the topsoil is well known. The
in conservation agriculture
economics of the system are discussed and its very positive
ecological effects are described at length.
The Brazilian experience
This publication is geared towards agronomists, advanced farmers,
extension workers and agricultural decision-makers throughout the
tropics and subtropics. It is hoped that the many lessons learned and
technologies developed in the Brazilian tropics can serve, with the
necessary local adaptation, as a starting reference for other tropical
(and subtropical) zones.

ISBN 978-92-5-105692-9 ISSN 1020-4555

9 789251 056929

Fao_Rom_01_10.qxp 24-04-2004 12:59 Pagina i

I n t e g r a t e d C r o p M a n a g e m e n t Vol. 5–2007

Tropical crop–livestock systems
in conservation agriculture
The Brazilian experience

John N. Landers
Associação de Plantio Direto no Cerrado, Brasília, DF, Brazil

Rome, 2007
Fao_Rom_01_10.qxp 24-04-2004 12:59 Pagina ii

The designations employed and the presentation of material in this information product do not imply
the expression of any opinion whatsoever on the part of the Food and Agriculture Organization of
the United Nations concerning the legal or development status of any country, territory, city or area
or of its authorities, or concerning the delimitation of its frontiers or boundaries.

ISBN 978-92-5-105692-9

All rights reserved. Reproduction and dissemination of material in this information product for
educational or other non-commercial purposes are authorized without any prior written permission
from the copyright holders provided the source is fully acknowledged. Reproduction of material in
this information product for resale or other commercial purposes is prohibited without written
permission of the copyright holders. Applications for such permission should be addressed to the
Chief, Electronic Publishing Policy and Support Branch, Communication Division, FAO, Viale delle
Terme di Caracalla, 00153 Rome, Italy or by e-mail to

Copies of FAO publications can be requested from:
Sales and Marketing Group
Communication Division
Food and Agriculture Organization of the United Nations
Viale delle Terme di Caracalla
00153 Rome, Italy
Fax: (+39) 06 570533360
Web site:

© FA O 2 0 0 7

Cover photos, from left to right: J. Landers, C. Batello, I. Döwich, J. Landers
Fao_Rom_01_10.qxp 24-04-2004 12:59 Pagina iii


Combining ecological sustainability and economic viability while
maintaining or improving agricultural productivity has long been a
matter of concern for FAO, as has reducing negative environmental
impacts. Conservation agriculture, which aims for zero tillage with the
maintenance of a surface mulch to protect the soil surface and increase
biological activity in the topsoil, is increasingly becoming recognized as
an effective system of crop production that protects the soil from erosion
while reducing the overall use of agrochemicals.
Vast areas of forest have been cleared in the tropical areas of Brazil for
establishment of pastures that become unproductive once the native
fertility of the soil is exhausted; this leads to yet more forest clearing for
new pastures. However, rotating pastures with field crops and resowing
is one of the most effective ways of maintaining them in a state of high
productivity, thereby reducing the need for more clearing.
This publication describes how pasture, fodder and livestock
production have been integrated into conservation agriculture systems
in Brazil’s tropical zones. Integrated crop–livestock zero tillage
systems (ICLZT) allow the sustainable production of high-yielding
pasture without further deforestation; in this system, grazing livestock
convert both pastures and crop residues into cash. The ability of
pasture to build up the biological activity and physical quality of the
soil is well known.
The lessons learned in Brazil by farmers and scientists can provide
valuable insights on what could be done in similar ecologies elsewhere,
including in Africa where the details of management will be different,
but the biological principles learned in Brazil could be a roadmap to
more sustainable intensification of some major crop–livestock
production systems.

Vol. 5–2007 iii
Fao_Rom_01_10.qxp 24-04-2004 12:59 Pagina iv


The manuscript was prepared by John N. Landers of Associação de
Plantio Direto no Cerrado, Brazil. It is hoped that this publication will
make the Brazilian experience in integrated crop–livestock zero tillage
systems accessible to agronomists, advanced farmers, extension workers
and agricultural decision-makers in other tropical and subtropical areas
so that they can adapt it to their own conditions.

Shivaji Pandey
Plant Production and Protection Division
FAO Agriculture Department

iv Integrated Crop Management
Fao_Rom_01_10.qxp 24-04-2004 12:59 Pagina v


The author would like to recognize the considerable assistance afforded
to the preparation of this manuscript by The World Wide Fund for
Nature (WWF) and The Nature Conservancy, Empresa Brasileira de
Pesquisa Agropecuária (EMBRAPA) research personnel at the Rice and
Beans, Beef Cattle, Soils, Maize and Sorghum and Cerrado centres, the
Instituto Agronômico de Paraná (IAPAR), Centre de Coopération
Internationale en Recherche Agronomique pour le Développement (CIRAD),
Luiz Carlos Balbino, Vilela and Cardoso and the farmer and agronomist
members of the Zero Tillage Farmers’ Association for the Cerrado Region
(APDC) who have collaborated with their experience and farm data.
Finally, and most importantly, great merit goes to FAO´s Plant
Production and Protection Division for recognizing the potential of this
new technology and giving financial support.
Thanks are especially due to the considerable input made by the
editors, in particular – retired staff members Jim M. Suttie and Robert
Brinkman, and Eric Kueneman, Caterina Batello and Theodor Friedrich
of the Crop and Grassland Service for ensuring that this manuscript was
brought to publication. Technical input from Josef Kienzle and further
editing by Stephen and Mary Reynolds are also appreciated. Finally, all
parties involved are very grateful to Dénise Welch, for her valuable inputs
and for revising and incorporating the many drafts into this final version.
The technical contribution of Paulo César de Faccio Carvalho,
Faculdade de Agronomia – Universidade Federal do Rio Grande do Sul
(UFRGS), Porto Alegre, Brazil has been much appreciated by FAO in
its effort to disseminate information on tropical crop–livestock systems
in conservation agriculture. He has also contributed, during his
assignment with FAO as a visiting scientist, to reinforcing information
knowledge management and partnership with EMBRAPA, UFRGS,

Vol. 5–2007 v
Fao_Rom_01_10.qxp 24-04-2004 12:59 Pagina vi


IAPAR and Universidade Federal do Paraná (UFPR) on crop–livestock
integration systems in Brazil.
The cover design and layout were done by Pietro Bartoleschi and his
studio in Rome. Chapter 1–4 photographs, unless otherwise
acknowledged, are by the author. Chapter 6 photographs came from
Theodor Friedrich.

vi Integrated Crop Management
Fao_Rom_01_10.qxp 24-04-2004 12:59 Pagina vii


iii Foreword
v Acknowledgements
ix Acronyms and abbreviations

1 Introduction
2 Conservation agriculture in the Brazilian tropics
3 Background
4 The Cerrado biome
5 The Amazon biome
7 History of zero tillage in the tropical zones of Brazil
7 Conservation agriculture
11 How does conservation agriculture work?
11 Integrated crop–livestock systems with zero tillage
14 Dissemination of ICLZT technology

15 Livestock and annual crop production in wet-dry and humid-tropical Brazil
15 Livestock type
16 Herd size and performance
17 Background for ICLZT
19 The process of pasture degradation

21 Principal integrated zero tillage crop–livestock systems
21 General considerations
26 Systems typology
27 Common rotations
28 Crop successions used as building blocks for rotations
28 Summaries of the ten main ICLZT technologies
28 1. Crop establishment in degraded pastures
29 2. Establishing pasture in annual crops
32 3. Sowing pasture after early harvest
33 4. Grass oversown in soybeans or maize
35 5. Grass regenerating during the first crop after ZT planting of a crop in old pasture
36 6. Planting forages on crop land for silage, green chop, dry season grazing or as a cover crop
36 7. Pasture renovation with forages sown jointly with grasses, for early grazing
38 8. Pigeon pea sown into existing pasture to improve winter grazing quality
38 9. Sowing perennial legumes into maize
40 10. Sowing soybeans in a permanent grass sward

Vol. 5–2007 vii
Fao_Rom_01_10.qxp 24-04-2004 12:59 Pagina viii


40 Opportunistic grazing of stubble in the dry season
40 Pigeon pea undersown in maize for stubble grazing
41 Grazing stubble in the dry season
41 Pasture grasses
43 Cover crops for grazing
44 Cut forage and silage crops
45 Pasture and grazing management
48 Legumes in pastures

49 Mechanized operations in zero tillage and soil fertility management
49 Residue management
49 Spraying desiccants and other chemicals
53 Planting and drilling
58 Soil fertility considerations

61 Technical and financial analysis of integrated crop–livestock zero tillage
61 Case Study 1 – A farm history of the adoption of CA with ZT
64 Without project
64 With ICLZT
65 Irrigated crop management – with and without project
66 Analysis of the Model Results
68 Case studies of other ICLZT technologies
70 Conclusions from the case studies

75 Sustainable agriculture and policy considerations
76 Farm-based economic benefits of CA, ZT and ICLZT
76 Farm-based environmental benefits of CA, ZT and ICLZT
79 Social benefits of ICLZT and increased land use intensity
80 Social support for conversion investments in ICLZT
81 Addressing the conversion needs of small farmers
84 Conclusion and policy recommendations
85 Policy recommendations for promoting ICLZT in tropical and subtropical regions


viii Integrated Crop Management
Fao_Rom_01_10.qxp 24-04-2004 12:59 Pagina ix


a. i. Active ingredient
ANA (Brazilian) National Water Agency
ANAE Association Nationale des Acteurs de L’Ecole
APDC Zero Tillage Farmers’ Association for the Cerrado Region
AU Animal Unit
CA Conservation agriculture
CEC Cation exchange capacity
CIMMYT International Maize and Wheat Improvement Center
CIRAD Centre de Coopération Internationale en Recherche
Agronomique pour le Développement
DAP Days after planting
DM Dry matter
DMP Deforestation mitigation potential
DOEN Dutch Foundation of Charitable Causes Lotteries in areas of
sustainable development, culture and welfare
EMBRAPA Empresa Brasileira de Pesquisa Agropecuária
FAO Food and Agriculture Organization of the United Nations
FEBRAPDP Federação Brasileira de Plantio Direto na Palha
GTZ German Agency for Technical Cooperation
GUS Groundwater ubiquity score
IAPAR Instituto Agronômico do Paraná
IBGE Instituto Brasileiro de Geografia e Estatística
ICLZT Integrated crop–livestock zero tillage system
INEMET Instituto Nacional de Meteorologia
INPE Instituto Nacional de Pesquisas Espaciais
INPEV Instituto Nacional de Processamento de Embalagens Vazias
IPM Integrated pest management
IRR Internal rate of return
IUCN World Conservation Union
LUI Land use intensity
NPV Net present value
SOM Soil organic matter
SSA Social Security Administration
TDN Total digestible nutrients (obsolescent energy standard)
UFPR Universidade Federal do Paraná
UFRGS Universidade Federal do Rio Grande do Sul
WWF World Wide Fund for Nature
ZT Zero tillage

Vol. 5–2007 ix
Fao_Rom_01_10.qxp 24-04-2004 12:59 Pagina x
Fao_01_14.qxp 24-04-2004 13:04 Pagina 1


Brazil is a world leader in conservation agriculture (CA); initially the
emphasis was on crop production through the now well-known
principles of maintenance of a layer of crop residues on the surface, zero
tillage (ZT) and crop rotations (see definition in Box 1). Now Brazil is
pioneering the integration of livestock production, grazed pasture and
forage crops into CA, grazing being so managed as to provide adequate
surface litter for the needs of ZT. A pasture phase in a rotation is
renowned for building up soil organic matter (SOM) and improving soil
structure. Pasture in ZT rotations with annual crops can be regenerated
much more profitably and with less risk than the older systems where
pastures were ploughed out before being resown.
This publication is aimed at agronomists, advanced farmers, extension
workers and agricultural decision-makers. It describes how CA has been
developed in the tropical and subtropical areas of Brazil and broadened
from a crop production system to an integrated crop–livestock way of
farming. Within Brazil there are many nuances in approaches adopted by
farmers and ranchers; there is not one fixed package for all situations.
However, the biology of the sustainable intensification made possible by
adoption of ZT has common principles.
Lessons learned in Brazil not only help Brazilian agriculture, but
provide insight on what farmers can do elsewhere. While the examples
described in Chapter 5 refer to large farms, many smallholders are now
applying ZT technology as well, giving proof that the principles are not
scale specific. Farming systems have to be developed to suit the agro-
ecological and economic characteristics of a region, but it is hoped that
many of the lessons learned and technologies developed in the Brazilian
tropics and subtropics can be adapted for use by farmers in other regions.

Vol. 5–2007 1
Fao_01_14.qxp 24-04-2004 13:04 Pagina 2


BOX 1. The definition of conservation agriculture (CA)

CA is a concept for resource-saving agricultural crop production that
strives to achieve acceptable profits together with high and sustained
production levels while concurrently conserving the environment. CA is
based on enhancing natural biological processes above and below the
ground. Interventions such as mechanical soil tillage are reduced to an
absolute minimum, and external inputs such as agrochemicals and
nutrients of mineral or organic origin are applied at an optimum level
and in a way and quantity that does not interfere with, or disrupt, the
biological processes.
CA is characterized by continuous minimum mechanical soil
disturbance (i.e. direct planting of crop seeds); permanent organic soil
cover, especially by crop residues and cover crops; and diversified crop
rotations in the case of annual crops or plant associations in the case of
perennial crops.

Source: Adapted from FAO, conservation agriculture homepage.

Broadly similar agro-ecological conditions to those in the Brazilian ZT
areas are widespread in parts of Africa and South and Southeast Asia, for
example (see maps in Nachtergaele and Brinkman, 1996).

This publication concentrates on integrated crop–livestock zero tillage
(ICLZT) technology which has developed principally in the Cerrado biome
(wet-dry tropical savannahs), extending to the Atlantic Forest biome (see
Figure 1). In principle, wider application is possible for the Amazon and the
Caatinga biomes, the latter representing the northeast semi-arid tropics.
There has been little development with ICLZT systems in the Caatinga, but
there is some research and farmer experience in the Amazon biome,
principally with soybeans and upland rice sown into degraded pastures.
This publication has six chapters. The introduction gives an overview
of CA, ZT and ICLZT systems before outlining their history and
development in the tropical areas of Brazil. The second Chapter
describes livestock and field crop production in the wet-dry and humid
tropical areas of Brazil and provides the background for ICLZT. The
third chapter gives the principles of the integrated systems and gives

2 Integrated Crop Management
Fao_01_14.qxp 24-04-2004 13:04 Pagina 3


Caatinga Biome

Amazon Biome Cerrado Biome

Pantanal Biome Atlantic Forest Biome

FIGURE 1: Location of the Cerrado, Amazon and other major biomes
Source: Instituto Brasileiro de Geografia e Estatística - IBGE, Brazil.

summaries of the ten main technologies as well as discussing fodder,
crops, stubble grazing, pasture crops, grazing management and the
grazing of cover crops. The fourth chapter discusses mechanized
operations and soil fertility management under ZT. The fifth provides
financial analyses and a series of case studies. The final chapter on
sustainable agriculture and policy considerations describes how land use
intensification can mitigate or obviate further forest clearing for pasture
establishment and deals with the positive environmental benefits of CA
and its impact on reducing the use of agrochemicals. The chapter ends
with policy recommendations for conversion incentives to ICLZT
technology, particularly for small farmers.

Agriculture has had different emphases in the Amazon and Cerrado
regions: beef, upland rice and small-scale dairying in the former;
large-scale beef and soybeans in the latter, with maize, upland rice and
cotton of secondary importance. Most clearing for pasture in the last
20 years has been on infertile soils, subject to rapidly falling stocking
rates as initial fertility declines and little or no fertilizer is used. This has
led to even more clearing to compensate for loss of carrying capacity.

Vol. 5–2007 3
Fao_01_14.qxp 24-04-2004 13:04 Pagina 4


About 55 percent of the Cerrado biome has been cleared (Machado et
al., 2004), as opposed to 16 percent of the Amazon biome (INPE, 2006).
Land prices have risen; cattle farmers are selling to crop growers and
moving to frontier areas to continue clearing. Land use intensification,
through integrating ZT crops with livestock systems is the most viable
route to preserve biodiversity on a large scale, provided that there are
policy incentives for it and disincentives to clearing new land.

The Cerrado biome
The descendants of seventeenth century mining pioneers grew subsistence
crops (on the less than 7 percent of the area which had fertile soils) and
reared beef cattle on the savannas, which were gradually impoverished
through burning at the end of every dry season to generate regrowth.
Improved pasture choices in the Cerrado were dictated by soil fertility.
On the few areas of fertile soils under forest (Nitisols) (IUSS, 2006),
mostly colonized by 1970 (Goedert, 1985), Molasses Grass (Melinis
minutiflora), Jaraguá (Hyparrhenia rufa) and to a lesser extent, Colonial
Guinea Grass (Panicum maximum) were sown. From 1972 these were
replaced by Brachiaria spp. as fertility dwindled, or the pastures were
renovated after a crop phase. In the latosols and quartz sands (Ferralsols
and Arenosols) (IUSS, 2006) of the Cerrado the opportunity to mine
native fertility by burning the sparse scrub is low, compared to Amazon
forest. Pastures were established after one or two pioneer crops of rice,
using very low fertilizer rates and minimal levels of lime, with little or no
maintenance fertilizer. Sano et al. (1999) estimated that over 70 percent of
Cerrado pastures were degraded.
Soybean production, the principal engine of development in the
Cerrado, began significant growth from 1980. The Cerrado, with its
gentle topography, wide interfluves, many tablelands, but essentially
infertile soils, was opened for grain crops by migrant farmers from the
southern states, mostly Rio Grande do Sul, using soil amendment
technology from the 1960s (McClung et al., 1958; Mikkelsen et al.,
1963). This technology was refined by the Brazilian Agricultural
Research Corporation (EMBRAPA), but only came into its own with
the advent of the first real tropical soybean cultivars “Cristalina” and
“Doko” in the late 1970s.

4 Integrated Crop Management
Fao_01_14.qxp 24-04-2004 13:04 Pagina 5


Early development of large-scale arable cropping was based on
subsidized agricultural credit. Conventional tillage was mainly with disc
harrows which caused disc pans and surface capping, and progressively
lower rainfall infiltration rates with erosion losses of up to 20 tonnes of
topsoil/ha/year (de Maria, 1999) and rapid depletion of the already low
soil organic matter, exacerbated by monocropping of soybeans. Contour
banks, required as a precondition for rural credit, were inadequate for
downpours of over 100 mm/hr and economic losses due to erosion
gradually worsened as soil structure deteriorated. Zero tillage came into
use when conventional tillage on infertile Cerrado soils was becoming
unattractive, since soil amendments amounted to about 25 percent of
direct costs and erosion losses were significant in terms of replanting and
lost soil amendments. Since about 1990 a significant number of crop
farmers have diversified into beef cattle, moving towards ICLZT systems,
as they acquired capital and diversified their risks, or bought degraded
pastures to expand soybean production.
On Cerrado latosols and quartz sands (Ferralsols and Arenosols),
(IUSS, 2006), improved pastures were based on Brachiaria decumbens
with some B. humidicola from 1972 onwards, usually undersown in
upland rice, with low initial fertilizer levels. The B. decumbens cultivars
(Basilisk and IRI 562) were replaced from about 1990 by B. brizantha cv.
Marandu and, to a lesser degree B. ruziziensis. Improved cultivars of
Andropogon gayanus, Paspalum atratum and Setaria anceps and attempts
to introduce pasture legumes had limited success. Predominantly
extractive management led to progressive pasture degradation (Macedo,
1997) and impoverishment of cattle farmers, who tend to sell land to crop
farmers and move to new frontiers. In rotation with crops, on amended
Cerrado soils, new, less stemmy and less clumpy Panicum maximum
cultivars (Tanzânia, Mombaça, Centenário and Vencedor) have
supplanted the old Guinea grass since 1990.

The Amazon biome
In the nineteen-seventies, farmers began to develop areas with fertile
soils in the Amazon in southern Pará, Acre, central Rondônia and
northern Tocantins states, where a forest-covered region with areas of
podsolic soils of some initial fertility (Acrisols) (IUSS, 2006), supported

Vol. 5–2007 5
Fao_01_14.qxp 24-04-2004 13:04 Pagina 6


Colonial Guinea Grass for up to 15 or 20 years. Guinea grass was
short-lived on poorer soils and was replaced by Brachiaria humidicola
and B. decumbens, and since about 1990, by B. brizantha, with a cycle
of up to 10 years before requiring renovation. Large areas of short-lived
Guinea grass, colonised by weeds and bush were abandoned due to
high renovation costs, especially where there were unrotted stumps.
Cattle expansion has pushed land clearing into rain forest on infertile
soils with a much shorter depletion cycle.
Pasture development was mostly by incomers from south Brazil,
spurred by subsidized credit (Mahar, 1989). Small farmers (local and
immigrant) in the Amazon opened forest land with subsistence crops,
intersown with pasture in the second or third year as increased weed
pressure made annual crops unprofitable. Pasture was established
cheaply between the stumps. Large farmers either bought land with
established pasture, or carried out more expensive mechanical clearing
and broadcast pasture seed after a burn. Pastures were established
without ploughing so there was some erosion control.
Where pastures are renovated with full land preparation, usually
undersown in rice, the only conservation measure, when present, is contour
banks, which can only be made once the stumps have rotted (about 20 years
after forest clearance). Contour banks are only effective so long as the soil
maintains a good infiltration capacity. This decreases after pasture renovation
due to compaction from cattle hooves and reduced ground cover, through
overgrazing, exacerbated by clumpy grasses like Colonial Guinea. Weeds
from South Brazil were introduced in dirty pasture seed. Residual effects
from former widespread use of Picloram herbicide for brush control in
annual cropping have not been assessed on a regional scale.
As pasture quality declined, migrant small farmers were forced to
intensify into milk production or depend on weaner-calf operations in
conjunction with extractive crops like cassava. Few had the capital to invest
in perennial crops on a significant scale, and extension services in the
Amazon and access to credit have historically been weak for this sector
(Camargo et al., 2002). Many farmers are in settlement projects with 100 ha
plots and have nearly exhausted the forest as a fertility source. Where small
farmers adopted conventional tillage it was almost invariably with
machinery hired from larger farmers, rarely employing contour banks.

6 Integrated Crop Management
Fao_01_14.qxp 24-04-2004 13:04 Pagina 7


Traditional populations used slash and burn on small 2–3 ha forest plots
with a long fallow and were close to sustainable, but with marginal income.

History of zero tillage in the tropical zones of Brazil
The first known tests with ZT in tropical Brazil were in Matão, São
Paulo state (Atlantic forest biome) in the 1960s, with a sod-seeder for
introducing legumes into pasture – however, this did not become general
practice (J. Harrington, personal communication, 1998). Herbert Bartz
began ZT on his farm in Rolândia, Paraná state, in 1972, just north of the
Tropic of Capricorn, and has used it continuously thereafter. His
experience is mainly suited to southern Brazil.
According to Landers (1994), in 1979 small farmers in Rondônia were
already jab planting beans (Phaseolus vulgaris) into rice straw after
desiccation with paraquat as a means of weed control. From 1981 onwards,
mechanized tropical ZT began to develop, with soybeans, maize and other
crops in the Cerrado, reaching over 9 million ha in the tropical areas in
2004–05, with some encroachment on the transition forest in Northern
Mato Grosso. Nearly all of this is in mechanized grain production. Today
ZT is being further developed in Roraima, Rondônia, Amazonas and Pará.
The advent of tropical ICLZT technology in the mid 1990s enabled the
sowing of grain crops directly into desiccated pastures, revolutionizing
crop and pasture production through a synergy which benefited both.
Since 1992 the Zero Tillage Farmers’ Association for the Cerrado Region
(APDC) has actively promoted ZT and, latterly, has been in the forefront
of promoting the ICLZT technology (Lara Cabezas and de Freitas, 1999;
Landers, 2001). Zero tillage is now the dominant production system for
annual crops in Brazil and is increasing rapidly in importance for pasture
renovation and establishment of perennial crops. ICLZT has shown
tremendous potential for land use intensification, reducing the rate of
clearing of native vegetation (see discussion in Chapter 6).

The term “conservation agriculture” was adopted during the First World
Congress on Conservation Agriculture, Madrid 2001, organized by FAO
and the European Conservation Agriculture Federation (Saturnino and
Landers, 2002).

Vol. 5–2007 7
Fao_01_14.qxp 24-04-2004 13:04 Pagina 8


Zero tillage is the preferred technology for CA1, the principles of
which are summarized in Box 2.

BOX 2. The 3 main principles of conservation
agriculture (CA)

1. Direct planting of crop seeds
2. Permanent organic soil cover
3. Crop rotation

Source: Adapted from FAO, conservation agriculture homepage.

The distinguishing and most important element of CA is that crop
residues remain on the surface, including those of cover crops (green
manures). The many functions of this surface organic layer, slowly
decomposing to humus, transform an untilled soil into a living, dynamic
system which must be managed to maximize these functions and derive
the benefits which they confer, resulting in increased SOM, greater P
availability and faster breakdown of agricultural chemicals. Conservation
agriculture replicates the closed cycle of nutrients and surface litter of a
mature rain forest, bringing highly productive farming into harmony
with nature. Water-holding capacity increases in proportion to SOM and
improved soil structure through an increase in water-stable aggregates
(Blancaneaux et al., 1993); water economy also improves since mulch
reduces evaporation losses. Stone and Moreira (1998) measured
reductions in irrigation demand of over 30 percent with an erect cultivar
of Phaseolus vulgaris planted in a thick mulch. The functions of surface
residues are shown in Box 3.
To plant in permanent residue cover it was necessary to develop
specialized planters and drills that cause minimal disturbance. These are
now available worldwide in manual, animal-drawn and mechanized
models (see Chapters 4 and 6). The sowing mechanism consists of a
In the pure interpretation of CA, tillage is not considered necessary for the creation of soil structure.
Therefore, CA encourages ZT, with the aim being no soil disturbance whatsoever. However, CA is also a
process with differing degrees of perfection, and in some cases a minimum disturbance of the soil may be
unavoidable. Examples of this include strip tillage in cold and wet climates, superficial ripping of the seedrow
after pasture to break up the compaction caused by animals, or even ripping in the row as long as the soil is
too degraded to maintain a structure (Friedrich, personal communication).

8 Integrated Crop Management
Fao_01_14.qxp 24-04-2004 13:04 Pagina 9


BOX 3. The functions of crop and weed residues kept
on the soil surface

• Weed suppression by shading or allelopathy;
• Retention of soil moisture through the mulch effect;
• Improvement of soil surface temperatures to within the limits for
biological activity;
• Reduction of the velocity of surface flow, reducing erosion potential
and increasing total rainfall infiltration;
• Elimination of the direct impact of rain drops on the soil, thus
preserving soil aggregates and avoiding surface capping;
• Nutrient recycling with slow liberation throughout the growth
period of the crop;
• Alteration of crop residue breakdown products through essentially
aerobic decomposition, in favour of the less acidifying humic acids
and complexing of aluminium in organic non-soluble forms;
• Alteration of the microclimate for emerging seedlings, protecting
them from wind-blown sand and excessive insolation (only applies to
standing stubble);
• Provision of organic substrates to feed soil flora and fauna;
• Generation of humus and other products of organic matter
decomposition which stabilise and increase organic carbon (Amado
and Costa, 2004) and improve soil structure;
• Food and shelter for wildlife.

Source: FAO, conservation agriculture homepage.

trash disc which cuts the crop residue and allows the fertilizer and seed
to be placed beneath it with minimum soil disturbance, leaving the soil
protected against rain and sun.
Multi-annual crop rotations, which are complementary to ZT in CA
systems, may include cover crops to build up residue levels, provide fodder,
and improve nutrient recycling from the subsoil. By not repeating the same
crop in the same season in successive years, the cycles of diseases, pests and
weeds are broken, promoting biological controls and reducing agricultural
chemical use and costs of production – this is discussed in Chapter 3.
Contrary to early predictions, once all plough-pans are removed (as
a pre-condition to adoption) and movement of lorries and heavy grain

Vol. 5–2007 9
Fao_01_14.qxp 24-04-2004 13:04 Pagina 10


trailers is restricted to roadways, ZT maintains soil structure at depth
by the preservation of macropores derived from old root holes and the
galleries and burrows of soil mesofauna, such as earthworms, beetles
and their larvae. Basic infiltration rates after two hours under these
conditions have been measured at 120 mm/hr, far higher than even
tropical rainfall intensity over the same period (Amado, 2005); without
such macropores this drops to 30 mm/hour or even 20 mm/hr when
soil is compacted. The macropores thus provide effective erosion
control under ZT; long term ZT farmers are removing their contour
banks because they no longer need them when they have infiltration
rates higher than rainfall intensity and a good residue cover protecting
the soil. An average of 30 erosion experiments in Brazil (De Maria,
1999) showed a reduction of 79 percent in soil losses under ZT
compared to conventional tillage (5.6 tonnes/ha/annum under ZT
and 23.3 tonnes/ha/annum with conventional tillage, while soil
regeneration is estimated at 10 tonnes/ha/annum).
The reduction in the negative impacts of erosion on-farm and off-farm
is, in itself, a huge environmental advantage. Farmers report increases in
terrestrial and aquatic fauna under CA. Aquifer recharge is discussed in
Chapter 6, which analyses the deforestation mitigation potential of land
use intensification with ICLZT. Perhaps the most impressive point about
farmers using CA is their change of attitude to working with nature
instead of against it (see Figure 2 below and more details in Chapter 6).




FIGURE 2: Zero tillage, the pathway to lower use of agricultural chemicals

10 Integrated Crop Management
Fao_01_14.qxp 24-04-2004 13:04 Pagina 11


PLATE 1: A trash disc cutting PLATE 2: Planting into a flattened cover of black oats (in
through crop residue – note the areas with cool winters). The tractor has double rear wheels
heavy spring to aid penetration. and 4 wheel-drive to reduce compaction. The cover after
planting is poor, but adequate to stop erosion.

How does conservation agriculture work?
The first operation of the agricultural year is desiccation of cover crops and/or
weeds with non-selective contact herbicides of negligible environmental
impact. Thereafter, specialized planters, or drills, cut the crop residue with
trash discs and slot the fertilizer and seed into the soil with minimal residue
disturbance (Plates 1 and 2). The only subsequent operations are application
of agricultural inputs (if required) and harvesting. Combine harvesters are
fitted with straw spreaders or choppers to give even cover of crop residues.
Conservation agriculture has engendered a more responsible attitude
towards the environment, regarding nature as an ally, and not as a foe. The
CA farmer combines crop rotations and maintenance of surface residue with
cover crops, integrated management of weeds, pests and diseases, rational
fertilizer practices, integration of crop and livestock enterprises, watershed
management and other environmental concerns. This confers sustainability
on the system and assures zero or low levels of chemical residues in
agricultural products, well within prescribed legal limits. Landers (2001)
estimated values for ZT and ICLZT environmental impacts in Brazil.

Zero tillage has facilitated the development of ICLZT in the Cerrado
biome, with wider application, in principle, in the Amazon, Atlantic
Forest and Caatinga biomes. Arable CA systems encompass a whole

Vol. 5–2007 11
Fao_01_14.qxp 24-04-2004 13:04 Pagina 12


farm, with both crops and livestock. Integration of beef fattening or
rearing with annual crops started as diversification under conventional
tillage. The new tropical ZT technology (Broch et al., 1997; Lara Cabezas
and de Freitas, 1999) eliminated the erosion risk from ploughing, gave
higher crop yields and tripled pasture carrying capacity while reducing
costs, as shown by Landers et al. (2005) in Chapter 5.
Farmers have encouraged technology development, facilitated trials on
their farms and often used technology ahead of research results. The
backup of wide-spectrum and interdisciplinary field research
(governmental and private) has resulted in credible long-term farm
results, specialised training and extension programmes.
A new era for tropical CA began in the 1990s, triggered by ICLZT,
which gave the potential to increase farm profitability and reduce
clearing of native vegetation through land use intensification. A crop
phase is the most cost-effective means of maintaining highly productive
pastures. Many cattle rearers lack cropping skills – in which case a long
term rental agreement with a crop farmer, guaranteeing winter grazing on
the stubble until the first crop land reverts to pasture, can satisfy both
sides, especially if Brachiaria spp. are undersown in the summer crop.
Crop farmers diversify into cattle more readily, not least because, in good
years, they have capital to invest and wish to spread their financial risk;
ranchers are more averse to ICLZT because of (generally unfounded)
fears of losing carrying capacity.
The principal benefits of adopting CA using ICLZT are summarized
in Box 4. At present, many of them are financial intangibles, but farmers
report them consistently as indirect financial benefits; direct financial
benefits are analysed in Chapter 5.
The gross margins at different levels of intensification of land use are
analysed in the real farm example shown in Figure 3. The benefits of
ICLZT are higher than for crop and livestock systems conducted
separately, illustrating that the benefits discussed above translate into
profits. In Figure 3, from left to right, beef cattle and crop enterprises
progress to an integrated crop livestock system on pasture and, finally, to
fattening yards with green chop and silage from an irrigated area. The
average of the first two bars is 26 percent; this should be compared with
the 46 percent of the third bar.

12 Integrated Crop Management
Fao_01_14.qxp 24-04-2004 13:04 Pagina 13


BOX 4. Benefits of adopting CA using ICLZT

1. Increased profits through reduced production costs;
2. Risk reductions through diversification;
3. Very low, or zero, pasture renovation costs;
4. Opportunities for strategic winter grazing on crop areas;
5. Increased whole-farm herd carrying capacity as a result of point 4;
6. Reduced disease, pest and weed pressures in crops;
7. Great reduction in environmental pollution through erosion control;
8. Consequent reductions in use of agricultural chemicals;
9. Maintenance of a high average stocking rate on rotated pastures;
10. Consequent mitigation of the demand for clearing new land;
11. Improved soil structure for annual crops through soil aggregation by grass roots;
12. Increased biomass generation for surface residues;
13. Increases in SOM, CEC and water-holding capacity;
14. Reduced fertilizer needs through recycling and reductions in leaching and
phosphorus fixation;
15. Improved rainfall infiltration rates, reducing erosion and flood peaks and
increasing aquifer recharge;
16. Reduced silting of reservoirs, especially those used for hydro-electricity
17. Faster depreciation of farm machinery;
18. Rationalization of overhead costs;
19. Professionalization of farmers and employees, with higher returns to labour;
20. Evening-out of income and labour peaks over the year;
21. Stabilization of the rural population and agribusiness job creation;
22. Generation of incremental employment in agro-industry.

Source: Adapted from Broch, personal communication (2002).

50 54%
40 46%
30 37%
10 15%
Beef cattle Annual ZT crops ZT crops/beef Intensive irrigated system
Solely beef cattle Solely crops Beef on pasture x crops Beef cattle in yards x cut forage from crop area
FIGURE 3: A comparison of gross margins at different levels of crop x livestock integration
Source: R. Merola, unpublished farm data.

Vol. 5–2007 13
Fao_01_14.qxp 24-04-2004 13:04 Pagina 14


Dissemination of ICLZT technology
A series of seminars, demonstrations, farmer technology meetings and
training courses involving ICLZT have been carried out since 1997,
mostly initiated by the private sector, more recently with active
participation of EMBRAPA, which now leads these efforts (Landers,
1999). State extension agencies, which primarily serve family farmers,
have been active in ICLZT in the last few years. In 2003 the Rice and
Beans centre of EMBRAPA published a complete manual of results on
ICLZT (Kluthcouski et al., 2003) and in 2005, the Ministry of
Agriculture, Livestock and Food Supply launched a programme
specifically to promote ZT, with training, demonstrations and research. In
2004 APDC initiated its “Stewards of Our Water” project with Petrobras
and in 2006 a joint APDC-National Water Agency (ANA) project was
launched, using ZT and ICLZT as the principal instruments to improve
on-farm water conservation. The “Pure Oil from Soya” project financed
by Stichting DOEN from the Netherlands aims to reduce chemical use in
soybean production. These projects obtain secondary financial support
from the private sector, state and municipal governments.

14 Integrated Crop Management
Fao_15_20.qxp 24-04-2004 13:05 Pagina 15

Livestock and annual crop
production in wet-dry
and humid-tropical Brazil

Beef cattle breeds in tropical Brazil are dominated by the Ongole zebu
from India, which has been the subject of selective breeding since the first
importations in 1870. The Nelore is the principal breed of Indian origin,
which has other minor derivatives (Oliveira, 2002). Other breeds from the
sub-continent are losing popularity as they are upgraded with superior
Nelore bulls. Bos taurus breeds dating from the colonisation period and
Santa Gertrudis (B. taurus x B. indicus) are of minor importance. European
beef breeds are not adapted to the tropics. For fattening, crosses of Nelore
or other zebus and European breeds are favoured; artificial insemination is
also used. Cows from these crosses come into heat at about 24 months, as
opposed to 30 months for pure Nelore (A. C. Campos, personal
communication, 2005). Magnabosco et al., (2002) demonstrated the effect
of genetic selection for performance on pasture: superior genotype animals
showed a 26 percent improvement in weight gain over poorer ones.
Most of Brazil’s milk comes from Holstein-Friesian cattle and their
crosses, of which the Girolanda is the most popular. At tropical
elevations over 600 metres, pure Holstein-Friesian perform well when
given optimal conditions of feed and shade. Progeny testing and selective
breeding programmes are advanced, as are specialised suppliers of semen
and embryo transplant services. Holsteins were predominantly of North
American origin, but recent imports of stock and semen from New
Zealand and other countries are genetic types selected for performance
on pasture. There are a few specialized Jersey herds.

Vol. 5–2007 15
Fao_15_20.qxp 24-04-2004 13:05 Pagina 16


Water buffalo thrive where they have access to water. Except for
Pará’s Amazon flood-lands the water buffalo population is low in the
rest of Brazil. They are now expanding from the Amazon lowlands to
the lower tropics, due to their hardiness and adaptation to poorer feed.
There is an excellent example of almost subtropical ICLZT with water
buffalo in Rolândia-PR, of many years standing, where biomass in
excess of 6 tonnes/ha is made into hay for winter feed on permanent
pasture (H. Bartz, personal communication, 1994).

Herd size and performance
Brazil’s cattle herd grew from 153 million in 1995–1996 to 205 million in
2004, an annual growth rate of 3.7 percent. Of these, 60 percent are in the
humid and wet-dry tropical states comprising the Amazon and Cerrado
biomes. The Cerrado figure is estimated since IBGE figures are by state,
not biome. The importance of the Cerrado in beef production is evident.
About 60 percent of Brazil’s 99 million ha of planted pastures are in wet-
dry and humid tropical regions: see Table 1.


Brazil 204 512 737 78 048 463 99 652 009
Cerrado* 84 596 782 17 443 641 45 320 271
Amazon** 39 787 138 9 623 763 14 762 858
* States of Mato Grosso, Mato Grosso do Sul, Goiás and Federal District, Bahia west of the São Francisco river,
plus 50 percent of Minas Gerais.
** States of Amazonas, Rondônia, Tocantins, Acre, Amapá, Pará and Roraima.

Source: Instituto Brasileiro de Geografia e Estatística - IBGE data for 2004 - website (2006).

Average performance parameters for Brazil’s cattle herd are given in
Table 2. The southern winter offsets the dry season in the Cerrado, and
most of Brazil’s herd is in these two regions, so these figures are considered
representative for the wet-dry and humid tropics. The livestock sector in
tropical Brazil is at the end of an extractive phase where the solution has
traditionally been to expand onto newly cleared land. From about 1980,
expansion was mainly on dystrophic latosols (Ferralsols) or quartz sands

16 Integrated Crop Management
Fao_15_20.qxp 24-04-2004 13:05 Pagina 17


(Arenosols), which required liming and had little inherent fertility.
Pasture establishment under rice, seldom using more than 300 kg/ha of
NPK 4.14.8, was common. Stocking rates fell fast and clearing continued
to compensate for both loss of carrying capacity and herd expansion.


Calving rate (%) 60 >70 >80
Mortality to weaning (%) 8 6 4
Weaning rate (%) 54 65 75
Post-weaning mortality (%) 4 3 2
Calving interval (months) 21 18 14
Age at slaughter (years) 4 3 2.5
Herd offtake (%) 17 20 22
Carcass weight (kg) 200 220 230
Kill out (%) 53 54 55
Stocking rate (AU/ha) 0.54 0.72 0.96
Liveweight gain (kg/ha/yr) 64 98 145
Weight gain (kg of beef/ha/yr) 34 53 80
Source: Macedo (1999).

Integration of beef fattening or rearing into annual crop systems began as a
hedge against low soybean prices, but with the advent (ca. 1994) of the new
tropical ZT technology (Broch et al., 1997; Lara Cabezas and de Freitas,
1999), its multiple benefits to both crops and pastures soon became evident.
Dry season pasture production determines a farm’s year-round
carrying capacity and beef prices rise in this period. In this publication
“carrying capacity” refers to the dry season stocking rate. Typical
rainfall patterns are shown for Amazon and Cerrado sites in Figure 4.
Rainfall intensity increases to the northwest and is greater in the large
cattle-growing area of South Pará, with a dry season of 2–3 months,
compared to 4–6 months in the Cerrado. Better winter rainfall in the
Amazon region allows higher dry season stocking rates and cattle
enterprises give a higher average return per hectare than in the degraded

Vol. 5–2007 17
Fao_15_20.qxp 24-04-2004 13:05 Pagina 18


400 400

300 300

200 200

100 100

0 0























Precipitation (mm) Precipitation (mm)

FIGURE 4: Monthly rainfall distribution at typical Cerrado and Amazon sites
Source: Instituto Nacional de Meteorologia - INEMET.

pastures of western São Paulo (models No. 4 and No. 5 respectively in
Chapter 5 and Camargo et al., 2002).
The author estimates that about 85 percent of pastures in the Cerrado
are Brachiaria with somewhat less in the Amazon biome, while Panicum
maximum cultivars make up less than 5 percent in the Cerrado.
Introduction of Brachiaria in the Cerrado raised average carrying capacity
from 0.3 AU/ha to 1.0 AU/ha (Macedo, 1997). Brachiaria pastures on
Cerrado latosols can last up to 5 years without further fertilizer; Guinea
grass cultivars give higher carrying capacity on more fertile soils, but
decline rapidly with loss of fertility. The winter carrying capacity of native
pasture is between 0.1–0.05 ha/AU with annual burning to generate
regrowth at the height of the dry season. Under this regime mortality was
high and steers lost weight badly over the dry season, reaching slaughter
(220–240 kg carcass weight) in three to four years.
Extensive beef producers have been slow to invest in management
technology, (e.g. fertilizing pastures, electric fencing, pasture legumes and
integrated crop–livestock rotations) but have rapidly taken up new grasses,
such as Brachiaria brizantha and new cultivars of Panicum maximum,
which show a very high benefit/cost ratio (Camargo et al., 2002). Extractive
management led to gradual decapitalization of beef farmers, which,
together with resistance to change, has impeded the modernization of the
industry. Growing maize and sorghum for silage is quite widespread and
dry season fattening pens have gradually increased since the early 1970s.
Since about 1990, there has been a significant diversification of crop
farmers into beef, moving towards ICLZT.

18 Integrated Crop Management
Fao_15_20.qxp 24-04-2004 13:05 Pagina 19


Vaccination against foot-and-mouth disease, control of ectoparasites
and ear tagging are obligatory in beef-exporting states and the rule in the
other states, except for ear tagging. Mechanical milking is increasing
rapidly, although most small producers still hand-milk; silage or green
chop are used extensively in winter and concentrate feed is based on
maize and soybean meal with additives. Small farmers are traditionally
milk producers on improved pastures; in the South of Pará there is a
concentration of small milk producers, currently in crisis due to falling
carrying capacities and lack of access to new land (Homma et al., 2001).
Available pasture technology permits profitable intensive cattle operations
(Yokoyama et al., 1999; Landers et al., 2005; and Vilela et al., 2001). Electric
fencing facilitates rotational grazing and the adoption of ICLZT is a low-cost
way to turn residual fertility from the crop phase into profit. Fertilizing
pastures is considered uneconomic at prevailing beef prices.

The process of pasture degradation
Inadequate replacement of nutrients and overstocking are the chief
causes of pasture degradation in the Cerrado (Vilela et al., 2004); Figure
5 shows the general sequence of degradation processes. Before going
into an ICLZT system the degree of pasture degradation must be taken
into account, to decide whether full land preparation is required. Besides
fertility decline and overstocking, brush invasion and large paddock size
are the major limitations to pasture productivity in the Amazon.


- N, - P, etc.
Loss of Productivity and Quality
PASTURE Weed Invasion


FIGURE 5: Schematic representation of the soil degradation process
Source : Macedo (1999) in Vilela et al. (2004).

Vol. 5–2007 19
Fao_15_20.qxp 24-04-2004 13:05 Pagina 20


PLATE 3: Light green/yellowish colour may be PLATE 4: An overgrazed Brachiaria pasture, with
an indication of N deficiency. low regrowth capacity, invaded by termite mounds.

Continuous degradation through extractive management (as shown in
Figure 5) brings successive reductions in carrying capacity. These begin
with N and P deficiencies, as shown in Plate 3, and end up in the loss of
soil cover and consequent soil degradation through compaction and
erosion. Overgrazing is the most common form of poor pasture
management. Ideal management seeks to keep the pasture with full ground
cover and always in the productive phase, by adjusting stocking rates or
rest periods to avoid over-grazing, and adding fertilizer and lime when
necessary. Short high stocking periods in the maintenance phase are
tolerable. An example of the end of the pasture degradation phase, about
to enter the soil degradation phase is shown in Plate 4, where annual beef
production would be about 40–50 kg/ha/year and dry season stocking rate
is below 0.5 AU/ha. Nitrogen deficiency is the principal limiting factor in
Brazilian tropical pastures (Boddey et al., 2004) followed by phosphorus,
sulphur, calcium, magnesium and zinc (Vilela et al., 2004).
In the Amazonian pasture areas of southern Pará the 2–3 month dry
season and better quality Guinea grass carried 1 to 2 AU/ha over the
first five years, depending on initial fertility. When it dropped to about
0.5 AU/ha, due to fertility decline or bush encroachment, the pasture
was renovated with Brachiaria spp., with carrying capacity ranging from
1.5 to 2 AU/ha in its first five years. Thus, B. brizantha or B. humidícola
replaced Guinea grass as fertility was mined, restoring profitability with
a second extractive phase (Kaimowitz and Angelsen, 2001).

20 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 21

Principal integrated
zero tillage crop–livestock

Zero tillage is not sustainable without crop rotation, always with a
different crop in the same season of the subsequent year. Rotations
comprise a number of single-year crop successions. In tropical Brazil a
second crop may be sown in the second half of the rainy season: where
rainfall is inadequate a fallow is used. Examples of typical rotations are
found below and others are detailed in the case studies (Chapter 5).
Soybean-millet is a common crop succession in one year, but
soybean-millet followed by soybean-millet is not a rotation, it is a crop
succession repeated. The greater the number and the higher the diversity
of crops and genera involved in a rotation, the higher the biodiversity
and the greater the potential for biological control of diseases, pests and
weeds, through cutting the build-up of inoculum or populations; even
nematodes can be efficiently controlled by some pasture grasses (Vilela
et al., 2004) and leguminous cover crops. A pasture phase in a rotation
builds up SOM, improves soil structure, nutrient and water availability
(Vilela et al., 2004), and allows reductions in the levels of pesticides and
fertilizers used. High SOM and crop residue levels reduce the potential
pollution of aquifers and surface water by acting as chemical and
physical buffers (Amado and Costa, 2004).
Crop successions within the rotation must be balanced so as to
maintain an average of over 6 tonnes/ha of dry matter in crop residues.
Production of adequate biomass to build up SOM and maintain
profitability is helped by a pasture phase. Cover crops can assist when

Vol. 5–2007 21
Fao_21_48.qxp 7-05-2004 15:30 Pagina 22


late or winter rains are adequate. Sá et al. (2004) demonstrated increases
of root mass of between 11 and 76 percent at 5 and 10 tonnes DM/ha of
residues with a positive regression on hybrid maize yields.
Table 3 shows the impact of the cover crop on the yield of the
following crop in Minas Gerais (the performance of sunflower depends
on low soil pathogen levels and absence of free aluminium) and Table 4
shows which main crops can follow which preceding crops from the
point of view of soil diseases and plant nutrition. The table also shows
recommended and unsuitable following crops, illustrating the principle
of rotating crop families. In Table 3, with results from Minas Gerais,
yields varied significantly for different preceding crops with both maize
and soya, with a general advantage for dicotyledonous preceding crops,
except for black oats (Avena strigosa) preceding soya.


Pigeon pea 6 628 a* Sunflower 3 558 a*
Niger oil** 6 367 ab Black oats 3 521 ab
Sunflower 6 195 bc Forage radish 3 440 ab
Pearl millet 5 981 bcd Sorghum 3 437 ab
Maize 5 964 cd Pearl millet 3 258 b
Sorghum 5 727 d Maize 3 002 c
* Letters indicate statistical differences at the 5% level (Tukey)
** Guizotia abyssinica
Source: Lara Cabezas, unpublished data (2003).

Table 5, with data from Roraima, 3° N, shows that soybean following
pigeon pea was very inferior to soybean following a graminaceous crop
(with or without another legume). This has been corroborated by
Döwich (2002) for Bahia and Kluthcouski et al.(2003) at several sites,
indicating the fundamental nature of this principle. The sources of these
yield increments are still to be elucidated.
Biological activity in the soil is increased under ZT. The most important
discovery in this area, using Brachiaria spp. as a cover crop, has been the
reduced incidence of soil pathogens, as shown by Da Costa and Rava
(2003), who measured reductions in Rhizoctonia (R. solani) and White

22 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 23



Not recommended Recommended Recommended Not recommended
forage radish, maize, sorghum, SOYBEANS maize, sorghum,
phaseolus bean, finger millet, finger millet,
pigeon pea, hairy wheat, rice, oats forage radish,
vetch, mucuna, lab (A. sativa), maize + sunflower, rice,
lab, trefoil, and mucuna and maize oats, hairy vetch
sunflower + phaseolus bean and wheat
sorghum, rice, finger soybeans, oats, MAIZE oats (A. sativa),
millet and wheat phaseolus bean, forage radish,
mucuna, maize, sunflower, finger
sunflower, sunn millet, wheat and
hemp, forage soybeans
radish, hairy vetch,
maize + mucuna
and lab lab
forage radish, bean, maize, COTTON oats, forage sunflower, phaseolus
hairy vetch, pigeon soybeans, radish, wheat, bean, hairy vetch and
pea finger millet soybeans, maize, trefoil
and oats sorghum, finger
millet and rice
forage radish, forage radish, S U N F L O W E R maize, sorghum, soybeans, cotton and
mucuna, pigeon pea, mucuna, pigeon rice, finger millet, phaseolus bean
trefoil, hairy vetch and pea, trefoil, hairy oats, wheat, and
lab lab vetch and lab lab forage radish
trefoil, pigeon pea, maize, sorghum, P H A S E O L U S maize, sorghum, forage radish,
soybeans, forage rice, oats, finger BEAN rice, finger millet, soybeans and
radish, lab lab, hairy millet and mucuna wheat and oats sunflower
vetch and sunflower
finger millet and rice maize, pigeon pea, SORGHUM sunflower, oats, wheat, maize,
soybeans, forage phaseolus bean, sorghum and finger
radish, crotalaria soybeans, trefoil, millet
and mucuna forage radish,
mucuna, hairy
vetch, pigeon pea
and lab lab

wheat, sorghum and forage radish, UPLAND sunflower, trefoil, wheat, maize,
finger millet crotalaria, pigeon RICE phaseolus bean, sorghum and finger
pea, mucuna, hairy vetch, oats, millet
trefoil, oats, maize, soybeans, forage
sunflower, lab lab radish, mucuna,
and phaselous pigeon pea and
bean lab lab

rice, sorghum, finger maize, crotalaria, W H E AT mucuna, sunflower, maize and rice
millet and black oats phaseolus bean, crotalaria, cotton,
for seed sunflower, pigeon phaseolus bean,
pea, mucuna, lab maize, sorghum,
lab, soybeans and finger millet, lab
cotton lab, pigeon pea
and soybeans

none any crop O AT S any crop wheat after black oats
to seed
cotton, sunflower, soybeans, maize, PERENNIAL soybeans and cotton, oats, rice,
wheat, sorghum and phaseolus bean, PA S T U R E phaseolus bean sunflower, maize,
upland rice oats and finger sorghum and wheat
Source : Broch et al. (1997).

Vol. 5–2007 23
Fao_21_48.qxp 7-05-2004 15:30 Pagina 24



Brachiaria spp. 3 775
Native grass + Stylosanthes guianensis cv. Lavrado 3 632
Stylosanthes guianensis cv. Lavrado 2 828
Pigeon pea 2 815
Source: Gianluppi (1999).

Mould (Sclerotinia) infections of 75 percent and also in Fusarium solani.
This principle has revolutionized rainfed and irrigated Phaseolus bean
production when planted into desiccated Brachiaria, sometimes leading to
farm yields of over 3 000 kg/ha (corroborated by Kluthcouski et al., 2003).
Dos Santos (2003) analysed different systems of ICLZT in terms of
weight gain (in beef) and financial returns to the beef enterprise (see
Figure 6). The introduction of a fast-establishing annual (hybrid forage
sorghum), gives grazing 20 days earlier compared to conventional
establishment of Panicum maximum. Alternatively, a mixture of pearl
millet (Pennisetum americanum), or finger millet, (Eleusine coracana)
with P. maximum can achieve the same result (A.B. van der Vinne,
personal communication, 2003).

800 868.15
400 478.87
300 361.77
Forage sorghum Forage sorghum Brachiaria
+ P. maximum + Brachiaria ruziziensis
cv. Mombaça brizantha (control)

Kg beef/ha/year

FIGURE 6: Performance of three systems of ICLZT in terms of weight gain and net profit in
Brazilian reais*
* US$ 1.00 = R$ 2.91, average for Nov. 2003 Source: Dos Santos (2003).

24 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 25


There is a direct correlation between the amount of crop residue dry
matter and the yield of the following crop, as shown by Séguy and
Bouzinac (2002). The farmer must know this to gauge the opportunity
cost of the pasture consumed (see Figure 7), but this requires a level of
cost analysis beyond that of most farmers. Pasture renovation with
ZT can be done without a crop phase, by desiccation of the old pasture,
re-seeding with fertilizer and liming, using part of the technology
described on the following pages.

7 000 1
( K g / h a )


6 000
5 000 5

4 000

3 000

2 000

1 000

0 2 000 4 000 6 000 8 000 10 000

D R Y M AT T E R O F C R O P R E S I D U E B I O M A S S ( K g / h a )

Regressions* of yield of a medium cycle soybean variety FT-114 on quantity and type of
biomass of the preceding crop. 1997/2000, Agronorte, Sinop, Mato Grosso, Brazil

1 Finger Millet - Ypg Ypg = 0,5957X + 948,76 R = 0,91

2 Pearl Millet - Ymi Ymi = 0,4528X + 2181,6 R = 0,90

3 Sorghum - Yso Yso = 0,629X - 154,63 R = 0,97

4 Sorghum + Brachiaria R. - Yso + B Yso + B = 0,4327X + 978,98 R = 0,80

5 Pearl Millet + Brachiaria R. - Ymi + B Ymi + B = 0,2787X + 2110 R = 0,70

Biomass (dry matter) production interval

* 6 replications x different fertilizer levels each year

FIGURE 7: Correlation between total annual dry matter generated by the preceding crop and the
yield of the following soybean crop
Source : Séguy and Bouzinac (2002).

Vol. 5–2007 25
Fao_21_48.qxp 7-05-2004 15:30 Pagina 26


Systems typology
Livestock systems in ICLZT are independent of the crops grown; the
main ones are:
(i) Fattening steers to slaughter weight;
(ii) Rearing operations selling steers to (i);
(iii) Combined rearing and fattening operations;
(iv) Cow calf operations selling weaners to (ii) and (iii);
(v) Breeding-to-slaughter integrated operations;
(vi) Intensive dairy operations;
(vii) Small, non-intensive, dairy operations.
All may or may not include supplementary feed with silage, cut fodder and
concentrates. In beef-from-pasture operations this is confined to the dry
season and creep (supplemental) feeding of calves at high management levels.
Use of salt or mineral mixes is general, even in the most extensive systems, and
mineral supplements usually include protein or urea to improve digestibility
of dry season forage. Feedlot fattening can be grafted on to (i), (iii) and (iv),
considerably increasing intensity of land use. Pure feedlot fattening is not
discussed here as it is an independent alternative to fattening on pasture.
There is no over-arching crop system typology. The chief distinction
is between rainfed and irrigated cropping, which is rarely without a
rainfed area. Few farmers change their basic cropping pattern when
adopting ICLZT, although they may undersow maize or oversow
soybeans with grasses or grow earlier varieties to allow greater winter
pasture production as a second crop. Some advanced farmers irrigate
intensively-grazed pasture. Cattle farmers adopting ZT cropping (a lesser
trend) usually choose soybeans or cotton as their main cash crop. The
latter is late-sown in December, which allows time for early bite grazing
at the onset of the rains with pearl millet, finger millet or forage sorghum.
Basic ICLZT systems have a crop and a pasture phase in rotation and
(i) Winter stubble grazing on summer cropland;
(ii) Summer crops with winter pastures (undersown or oversown);
(iii) Summer crop plus second crop plus stubble grazing in winter;
(iv) Crop production for feed supplement (silage, sugar cane, elephant
grass, hay, green forage), usually a minor area within another system;
(v) Some combination of these.

26 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 27


Common rotations
Soybeans are often mono-cropped for several years during the low
disease-weed-pest pressure period after land clearing and are
universally dominant in rainfed farming, with maize in rotation, when
practised, every two to four years. Soybeans do not return enough
biomass to the system for long term sustainability. While physical soil
degradation has been severe under conventional soybean cultivation it
has been hardly noticeable under ZT, leading many farmers to disregard
a pillar of CA, crop rotation.
Maize is essential for adequate biomass generation in a CA rotation,
but in frontier areas the market is weak; higher transport costs per unit
value and higher storage requirements (both double those of soybeans)
severely limit its use in rotations. In such regions rice is often grown as
a pioneer crop after clearing (Séguy et al., 1998), or in pasture renovation
(Kluthcouski et al., 1999). New rice cultivars can exceed 5 tonnes/ha
(Séguy et al., 1998). Cotton and Phaseolus beans have the advantage of
late sowing, allowing more grazing on winter pasture. In the western
Cerrado and in Amazonia a second crop of maize can be highly
profitable. Alternatives to maize as a second crop in lower rainfall areas
are pearl millet (Pennisetum americanum), sorghum (Sorghum bicolor)
and finger millet (Eleusine coracana), or pasture grasses sown in
association with the main or second crop.
A common starting point for adoption of ICLZT would be a 10 to
20 year-old degraded pasture. The most generalized ICLZT rotation would
be: Yrs 1, 2 and 3 soybeans in summer and stubble grazing in winter;
Yr 4. maize, undersown with Brachiaria brizantha or B. ruziziensis; years
5–7 or 5–8 pasture.
The most sophisticated rotation would be: Yr 1 cotton-winter fallow,
Yr 2 soybeans-second crop maize, Yr 3 soybeans-second crop maize
undersown with finger millet and Panicum maximum, Yr 4 one-year
P. maximum ley. This maximises the stocking rate on the pasture up to
4–5 AU/ha, resulting in 75 percent of the farm in crops (Vinne, 2001).
The simplest rotation would be that of Case Study 5, a predominantly
cattle operation, with: 6 year-old pasture converted to Yr 1 maize for
silage and/or grain, undersown with grass, Yr 2–7 pasture (enriched, or
not, with pigeon pea).

Vol. 5–2007 27
Fao_21_48.qxp 7-05-2004 15:30 Pagina 28


Crop successions used as building blocks for rotations
Crop successions, which are the building blocks for any desired rotation,
can be classified as follows:
• Crop establishment in degraded pastures;
• Pasture establishment in, or immediately following, an annual crop
(for permanent or temporary pasture). This could include dual use
of a cover crop both for grazing and subsequent soil cover;
• Use of crop areas for silage or green chop production as supplement
for cattle;
• Enrichment of existing pastures with legumes. This is practically
limited to sowing pigeon pea directly into existing pasture;
• Sowing annual crops into a permanent grass or legume sward;
• Opportunity grazing of crop stubble in the dry season.

1. Crop establishment in degraded pastures
Excessive impediments to mechanized cultivation, or toxic aluminium
levels in the 0–20 cm layer, may require full land preparation to remove
irregularities and incorporate lime. When necessary, operations should be
carried out towards the end of the rains and an annual fodder sown
immediately to provide enough winter grazing to more than compensate
for loss of the old pasture. This removes the chief concern of the rancher
(i.e. lack of winter forage). These fodders must be grazed to suppress
flowering or they will mature and die back. If they set seed, it will fall to
the ground and germinate with the first rains. Either from regrowth or
seed, ground cover can be generated and ready for desiccation prior to
sowing the annual crop (see Box 5). In the case of subsoil deficiency in
calcium or sulphur deficiency, gypsum can be applied along with lime,
whether incorporated or not.
If no impediments exist the degraded pasture is desiccated at the
beginning of the rains with a systemic, non-selective herbicide
(glyphosate, or ammonium gluphosinate). Specialized ZT planters (Plates
19, 21 and 23–25) place seed and fertilizer below the residue. A narrow
in-row tine or special fertilizer boot behind the cutting disc (Plates 26 and
27) is necessary in old, degraded pasture to break the compacted surface
layer, allowing deep fertilizer placement and penetration of crop roots.

28 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 29


BOX 5. Critical points: Crop establishment in degraded
(i) Pasture should be heavily grazed or cut prior to desiccation, which is
done when leaf area has recuperated enough to absorb the herbicide;
(ii) Clumps (tussocks), if not well rotted, will cause the planter to jump
and leave seed exposed on the surface even after rotary chopping;
(iii) An interval of up to 21 days is needed after desiccation to allow
root fermentation – a light second desiccation is usually required
just prior to sowing.

Where there is no compacted layer or excessive residue to clog the tines,
offset double discs of different diameters for both seed and fertilizer are
used (Plates 26 and 27; more information in
Baker, 2007). Thereafter selective herbicides
are used ad hoc in the crop. A one-year
pasture ley maximizes the proportion of
crops in the rotation for a given herd size,
which is an option when crop prices are high.
When soil conditions are adequate any PLATE 5: Adhemar Tietze, pioneer
guacho farmer in Sinop, Mato
annual crop can be sown with this technology Grosso state, shows desiccated
Brachiaria residue in ZT soybeans.
– soybeans (see Plate 5), phaseolus beans,
maize, cotton etc. There is a special bonus with phaseolus beans because
of the low level of soil disease and nematodes after desiccating Brachiaria
(Da Costa and Rava, 2003; and Vilela et al., 2004).

2. Establishing pasture in annual crops
The most common options for establishing pasture in annual crops are:
undersowing (see Plate 6), intersowing and oversowing.
Rainfall availability is greatest with undersowing and least with
sowing after harvest. CAT Uberlândia (2005) has shown no difference
in maize yield when 9 kg/ha of Brachiaria ruziziensis seed was sown at
the two and four leaf stages of maize, but the two leaf sowing gave
quicker establishment and allowed earlier grazing. Kluthcouski et al.
(2003) affirm that Brachiaria undersown in maize can be shaded to the

Vol. 5–2007 29
Fao_21_48.qxp 7-05-2004 15:30 Pagina 30


PLATE 6: Nelore cattle on , established under maize.
B. brizantha

point of not requiring herbicide to check it (tropical grasses are C4
plants, hence shade intolerant). For manual systems, a jab planter can
be used, mixing seed with fertilizer or some
inert granular material, but if the stand is
not uniform shading will be inadequate and
some yield loss will occur from competition
from Brachiaria in the unshaded areas.
Oversowing can gain up to three weeks on
same-day sowing after harvest, which is
crucial because the amount of rainfall after
germination is correlated with establishment
and dry matter yield. However, conditions
for establishment with oversowing are not PLATE 7: spp. intersown

in maize at 20 DAP.
always adequate and this risk must be
accepted against the potential benefit and low cost factors. Intersowing is
carried out by planting in the inter-row of the crop after emergence (see
Plate 7 and also Plate 8).

Maize undersown or intersown with Brachiaria is the most important
method for establishing pastures (see Box 6 for more details), and may have
nurse crops other than maize e.g. rice (Kluthcouski et al., 2003). Grass may
be undersown, mixed with fertilizer at sowing, dropped onto the row in
front of the trash disc from additional seed hoppers or by shallow drilling
two rows in the inter-row at the 2–4 leaf stage of the maize. Excessive

30 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 31


PLATE 8: Brachiaria intersown in maize; note the height of the grass shortly before maize harvesting.

BOX 6. Critical points: Establishing pasture in
annual crops
(i) Brachiaria seed mixed with fertilizer should be sown in maize
7–10 cm deep in lighter soils and 3–4 cm deep in heavier ones to
delay emergence (Cobucci and Portela, 2003); some farmers have
reported up to 10–20 percent yield depression in very wet years
(about 1 in 10 frequency) with this method, using B. brizantha, but
little with the less aggressive B. ruziziensis;
(ii) Seed mixed with fertilizer should be sown the same day or vigour
may be affected, especially with fertilizers containing high K or N
levels (Kluthcouski et al. 2003). Experience indicates a possible
problem of seed burn if, in a dry spell, the soil has inadequate
moisture to dilute the fertilizer;
(iii) Unless the soil is very well structured, the ZT planter or drill must
be equipped with a narrow in-row sub-soiling tine to break the
5–10 cm compacted zone in the old pasture – alternatively, larger
diameter fluted trash discs might be tried;
(iv) Panicums and other small-seeded grasses are unsuited to deep
planting and must be shallow-drilled, about 1 cm; at present
insufficient information is available to define the parameters for
undersowing small-seeded grasses in general.

Vol. 5–2007 31
Fao_21_48.qxp 7-05-2004 15:30 Pagina 32


grass growth can be checked with sublethal doses of Nicosulfuron of
8–12 gm a. i./ha (Cobucci and Portela, 2003). Experiments in several
locations and years showed little or no significant difference in maize
or sorghum yields (Kluthcouski et al., 2003), see Table 6.

In upland rice, undersowing has been common under conventional tillage
and works well with ZT when the grass is sown in the rice row to control
shading. Although some yield reduction of rice has been observed
(Kluthcouski et al., 2003) – the same depths of sowing should be observed
as for maize. undersowing is possible in soybeans: reduced post-emergent
applications of haloxyfop-methyl are promising and Cobucci and Portela
(2003) recommend one quarter of the full application to check Brachiaria
spp., but this technology is not well field-tested and results will be very
dependent on the metabolic activity of the grass.

3. Sowing pasture after early harvest
Usually, but not always in the Cerrado, there will be sufficient rain to
establish a pasture up to late February (see Box 7). The ideal method for
sowing is with a ZT drill in closely-spaced rows (approx. 15 cm), straight
after harvest (i.e. behind the combine on the same day). Drilling is
preferred, but most farmers do not have a drill and broadcast the seed with


CROP YIELD (kg/ha) 1
Maize 2 and 3
6 354– 6 877 6 401– 6 795
Forage maize 2
48 367 48 467
Grain sorghum 4
3 687 3 581
Forage sorghum 4
32 333 32 867
Soybeans 2 and 5
2 971– 3 056 2 677– 2 414
Rice 6 and 7
1 968– 2 072 1 503– 1 859
Average of six repetitions;
Average of four locations;
Average of two locations: Santa Helena de Goiás-GO e Luziânia-GO.
Refers to application of 6g of a. i./ha of nicosulfuron in maize undersown with grass;
Average of two locations: Santa Helena de Goiás-GO, Luziânia-GO and Mimoso-BA;
Average of three locations: Santa Helena de Goiás-GO, Campo Novo dos Parecis-MT and Mimoso-BA. Refer to
application of 24 g of a. i./ha of haloxyfop-methyl in soybeans undersown with grass;
Average of two locations: Luziânia-GO and Mimoso-BA;
Average of Luziânia-GO. Refers to application of 120g of a. i./ha of clefoxydin in rice undersown with grass
Source : Adapted from Kluthcouski et. al. (2003).

32 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 33


BOX 7. Critical points: Sowing pasture after early
(i) Adequate rainfall to get through the dry season should be
guaranteed by early sowing;
(ii) Local trials are necessary to optimise sowing date with variations in
rainfall, soils, varieties, etc.

a fertilizer spreader, covering it by running a closed levelling disc harrow
over the stubble. This low-cost method leaves about 40–60 percent cover,
but it oxidises organic matter and increases the evaporation of precious
moisture. In addition, it only has a moderate chance of success after
February (i.e. it is best after an early harvest, but then the farmer’s choice
would be for a second cash crop, so pasture establishment with this
system will always be in moisture-short situations). Finally, care must be
taken to minimize soil and residue disturbance – ideally, over 70 percent
cover should be maintained to guarantee erosion control.
An innovative São Paulo farmer’s solution to this problem was the
adaptation of a levelling disc, welding angle iron bars across old, worn,
discs, which leaves the cover practically intact. A chain harrow, or light
anchor chain, between two tractors, are superior solutions and faster
operations, incorporating the broadcast seed into the crop residue with
minimal disturbance. Desiccation may be unnecessary if the weed burden
is very low. Previous history of the field, plus spot observations should
determine this option – persistent weeds, such as Sida spp., Cassia
rotundifolia, Commelina spp. and Digitaria insularis should not be
present. As much starter fertilizer at planting as the farmer can afford
(emphasizing P) will assist in promoting good weed suppression and
providing early bite. The amount of winter pasture in the first year will
be governed by rainfall and overgrazing should be avoided.

4. Grass oversown in soybeans or maize
Oversowing can be done with a fertilizer broadcaster or by aerial
seeding after field-testing to check the required overlap. Commonly-
used forages in soybeans at full grain maturity are: pearl and finger

Vol. 5–2007 33
Fao_21_48.qxp 7-05-2004 15:30 Pagina 34


millets and Brachiaria spp., but Guinea and Molasses grasses have also
shown some success (Landers, 1994). This system is cheap (just the seed
plus application), offset by high risk of failure if rain does not follow
shortly after seeding; the Cerrado and Amazon transition forest regions
experience short dry spells of ten days or more during January-March
(Goedert, 1985) when soybeans are maturing. Oversowing into maize
increases the risk since there is no leaf-fall to cover the seed; it should
be carried out at the beginning of canopy senescence, when more light
penetrates to ground level. With any main crop, the earlier it is sown
and the shorter its cycle, the greater amount of rain will fall on the
second (oversown) crop (see Box 8). Rarely will the conditions exist for
oversowing of the whole cropped area. However, the cash value of
additional dry season grazing is very high. Beware of ants, beetles,
other insects or birds reducing establishment by removing seed – an
increase in seed rate may be required. Seed dressing with insecticide in
this sowing mode is dangerous for birds and small animals, and should
be avoided. However, fungicidal dressing can be useful to combat high
pathogen pressure in the surface litter. B. ruziziensis germinates better
on the soil surface than B. brizantha.

BOX 8. Critical points: Grass oversown in soybeans or
(i) The best chance of success is when soil is moist at sowing and rain
falls within a few days;
(ii) The risk/benefit ratio must be clearly understood, or an eventual
failure will lead to abandoning this practice;
(iii) Seed rates must be 2.5 or 3 times higher than for drilling – for seed
rates with drilling see Table 7;
(iv) The best time to oversow in soya is between seed maturity (when
the green grain comes easily away from the pod) and beginning of
leaf fall and, under favourable circumstances, with at least light
rain, up to 50 percent leaf fall (Landers, 1994);
(v) Pelleting to improve germination potential has not been promising,
but new hygroscopic materials would be superior to rock phosphate
pelleting (W. Lara Cabezas, personal communication, 2003).

34 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 35



GRASS SPECIES (kg/ha PLS)* (No.)
Andropogon gayanus 2.5 360
Brachiaria brizantha 2.8 150
Brachiaria decumbens 1.8 200
Brachiaria humidicola 2.5 270
Brachiaria ruziziensis 2.0 230
Paspalum guenoarum 1.5 300
Paspalum notatum cv. Pensacola 1.5 610
Panicum maximum cv. Tanzânia 1 1.6 960
Panicum maximum cv. Tobiatã 2.5 680
Panicum maximum 1.2 1 900
Setaria anceps cv. Kazungula 1.2 1 490
* Pure live seed (= germination % x purity %).
Source: Adapted from Kluthcouski et. al. (2003).

5. Grass regenerating during the first crop after ZT planting
of a crop in old pasture
As the crop canopy thins, seeds will germinate from the surface seed
bank (see Box 9). Under rainfed conditions Brachiaria spp. will not be
daylength-triggered to flower before it is time to desiccate for a second

BOX 9. Critical points: Pasture regenerating from
surface seed during the first crop after ZT
planting of a crop in old pasture
(i) It is important to achieve a full canopy quickly to suppress any grass
which germinates;
(ii) Graminicides should not be used in the summer crop, except at
sub-lethal doses, to check competition from the grass, if necessary
(Cobucci and Portela, 2003); this may
have variable results due to differential
response at varying levels of metabolic
(iii) Longer cycle crops or harvest delay
due to rain may cause excessive grass
growth which hampers harvest –
maximum combine speed is achieved PLATE 9: Removal of row separator
points reduces the quantity of
by removal of the row divider points, grass passing through the combine
and permits high speed harvesting.
except the lateral ones (see Plate 9).

Vol. 5–2007 35
Fao_21_48.qxp 7-05-2004 15:30 Pagina 36


crop, so the seed bank will dwindle. The decision to leave for winter
pasture or desiccate post-harvest with a half dose of systemic desiccant
will depend on the grass stand and the weed load. Dense stands of
Brachiaria are good at weed suppression.

6. Planting forages on crop land for silage, green chop, dry
season grazing or as a cover crop
Forage sorghum, pearl millet or finger millet can be drilled with ZT as a
second rainfed crop for green chop forage, silage, grazing or cover crop
biomass (see box 10). Sowing in January, or first half of February, will
maximise rainfed dry-matter production, but this is seldom done, since
the forage would compete with an economic second crop. Drought-
resistant crops are usually sown after March 1st. Drilling is the preferred
method for millets. Because of high residual nutrient levels from the
preceding crop, little, if any, fertilizer is used.
Irrigated silage and green chop, if possible, should be scheduled away
from competing with beans or other high value crops in the dry winter,
when these are grown – there is inevitably a conflict with green-chop for
fattening yards whose demand is constant. Seed rates for forage sorghum
are between 8 and 12 kg/ha in 50–70 cm rows, for pearl millet 15–20 kg/ha
and finger millet 10–15 kg/ha, both in 15–25 cm rows. Seed depth for
sorghum should be 3–4 cm and <2 cm for millet. In high fertility
conditions under irrigation, for silage production, specialised sorghum
hybrids can out-yield maize (up to 70 tonnes of silage/ha), at reduced
fertilizer rates, while maize reaches about 45 tons/ha under the same
conditions (farm data, S. Conrado, Cristalina – GO 2003).
Some intensive systems have been producing two crops plus dry-
season grazing: Séguy and Bouzinac (1997) report soybean yields of
4 000–4 600 kg/ha, 1 500–3 500 kg/ha of second crop grain plus 1–1.5 AU/ha
for 90 days in the dry season with a mixture of the three forages. This
gave a total gross margin for the annual succession of US$ 150–350/ha.

7. Pasture renovation with forages sown jointly with grasses,
for early grazing
A specially-designed ZT combination drill has three hoppers (grass
seed, forage seed and fertilizer). A set of trash discs cut through crop

36 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 37


BOX 10. Critical points: Use of crop land for silage or
green chop as supplement to cattle, for dry
season grazing or a cover crop
1. Rainfed areas:
(i) Taking maize silage may allow a second crop to follow or, with
sorghum x Sudan grass hybrids, a second cut;
(ii) Harvesting may leave ruts or compaction which require conventional
tillage to remove, but where Brachiaria is undersown, its root
development can counter the compaction; planting the silage/green
chop crop on desiccated pasture would be ideal in this respect;
(iii) Regrowth from sorghum after the last cut can give valuable dry
season grazing or biomass as a cover crop, to be desiccated prior to
the next crop. Note: in frost-prone areas, cattle must be removed
after frost, to avoid possible hydrocyanic acid poisoning;
(iv) For grazing, these forages should not be allowed to initiate
flowering – this will maximise regrowth and tillering.
2. Irrigated areas:
(i) Silage and green chop are best cut in the drier months of April to
September (silage to June only), to avoid soil compaction and
formation of ruts;
(ii) Phaseolus beans are the prime irrigated crop and are grown from
May to July (90 day cycle); rotation with a grass (especially Brachiaria)
is highly beneficial for disease control (Nasser et al., 2001; Da Costa
and Rava, 2003) and soil structure improvement for the bean crop.
Light grazing may be started as early as 30 days, but on loose soils
care must be taken that the cattle do not uproot the plants.

residue, followed by a narrow fertilizer boot with scarifier tine in the
row (to cut through the 5–10 cm surface-compacted zone in the old
pasture and place the fertilizer about 15 cm

deep), as shown in Plate 10 (detailed view in
Plates 26 and 27). This is followed by
alternate sets of offset double disc furrow
openers with different size discs for the
grass and the forage sorghum, respectively,
at about 15–18 cm spacing with open clod-
PLATE 10: A seeder-scarifier for
busting rollers behind. sod seeding.

Vol. 5–2007 37
Fao_21_48.qxp 7-05-2004 15:30 Pagina 38


8. Pigeon pea sown into existing pasture to improve winter
grazing quality
For sowing pigeon pea into existing pasture (see Box 11), the seeder-scarifier
(Plate 10) is used or a normal ZT planter is used with a fertilizer knife to
break the surface-compacted layer, and not a double disc. Fertilizer
(especially phosphate) applied in planting will also renovate the grass and the
pigeon pea can fix up to 150 kg/ha of N, also improving grass vigour (Dos
Santos, 2003). The extra protein of the legume allows cattle to utilize poor
quality winter grass efficiently. Pigeon pea seed rate should be 50 kg/ha and
planting depth 4–5 cm. This technology can be adapted for Leucaena
leucocephala and other legumes, although there is little local experience.
Leucaena suffers from severe leaf drop in the dry season, and did not fulfil
early promise (C. U. Magnabosco, personal communication, 2003).

BOX 11. Critical points: Pigeon pea sown into existing
pasture to improve winter grazing quality
(i) The lime and aluminium status of the soil should be adequate for
pigeon pea, otherwise liming is required;
(ii) Lime should be surface applied, with, at the very least, four months
of rain prior to sowing;
(iii) Cattle trails should be scarified, termite mounds and erosion gullies
(iv) Grazing management should allow time for the pigeon pea to
(v) A pigeon pea variety should be selected with characteristics similar
to the Brazilian “Bonamigo Super N”, which is usually avoided by
the cattle until the grass protein content falls below the critical 6
percent for maximum rumen activity;
(vi) On many soils, the pigeon pea nodulates with native rhizobia,
however performance may be enhanced by inoculating with a
specific strain.

9. Sowing perennial legumes into maize
This technology has good potential, but requires more development of
control mechanisms to check established perennial legumes (sub-lethal
doses of selective herbicides, or for small farmers, mechanical treatments);

38 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 39


(see Box 12). A forage legume is established
by undersowing in maize either using a
pelleted insecticide applicator coupled
behind the seed discs or by mixing the seed
with phosphatic fertilizer and directing the
fertilizer spout towards the soil falling back
in the furrow after the seed disc (Landers,
1994). This implies pre-planting fertilization
for the maize and topdressing of K and N, or
a combination planter with three hoppers.
The most compatible legumes were Siratro
(Macroptilium atropurpureum) (see Plate 11),
followed by centro (Centrosema pubescens), PLATE 11: Siratro (Macroptilium
atropurpureum), a deep-rooted
but kudzu (Pueraria phaseoloides) and glycine perennial forage legume, is very
compatible for sowing into maize.
(Neonotonia wightii) were more competitive.

BOX 12. Critical points: Sowing perennial legumes into

(i) A selective herbicide compatible with both maize and legumes
(e.g. alachlor), can be used in the establishment year to control
weeds at normal dosage rates for maize;
(ii) 2,4-D is not effective on vigorously growing perennial soybean and
some Stylos, while some maize hybrids may be susceptible to stalk
(iii) Local tests are required to find the best companion legume – non-
climbing ones might be better e.g. Stylosanthes spp., but not
Arachis pintoi, which is too vigorous;
(iv) The legume may be a bit weedy in the establishment year, but
after that it comes away fast after the first rains and smothers
annual weeds;
(v) Care must be taken not to allow the legume to compete excessively
with the maize, when severe yield depression will result;
(vi) Legume tendrils on the maize may slow combine speed for grain
harvest but do not affect silage choppers, and the legume
enhances the quality of the silage.

Vol. 5–2007 39
Fao_21_48.qxp 7-05-2004 15:30 Pagina 40


For silage, this system is excellent, leaving a full soil cover and giving
high quality winter grazing which can be used as a protein bank. Also,
the legume and surface litter will reduce surface compaction from
machinery. A double straight disc between the rows, at about 10 cm from
the maize row, can cut the tendrils and prevent the legume from climbing
on the maize.

10. Sowing soybeans in a permanent grass sward
This technology has been tested in Mato Grosso (Séguy and Bouzinac,
2002); the idea is not to kill the grass (see Box 13), but to arrest growth
up to soybean grain maturity. A Tifton cultivar of Cynodon dactylon
was used with reported soybean yields of over 3 000 kg/ha. The grass,
which can be grazed in the dry season, helps to suppress annual weeds
and provides biomass for soil cover. The same authors obtained 90 days
winter grazing for 1–1.5 AU/ha with this system and a total gross
margin of US$ 200–400 per ha for the annual succession.

BOX 13. Critical points: Planting soybeans in a
permanent grass sward
(i) Correct sub-lethal doses of graminicides may be necessary to check
grass growth; effectiveness will depend on the physiological
activity of the grass;
(ii) The compatibility of other crops with this system needs to be
tested, in order to obtain a sound rotation.

Pigeon pea undersown in maize for stubble grazing
This is a traditional practice reported by Von Schaffhausen (1949). Pigeon
pea seed is mixed with the fertilizer and deep sowing (10–15 cm) delays
its emergence in the maize row, where it remains shaded until maize
senescence (see Box 14). Manual jab planting in the row would follow
planting of maize after 30 days. In the second year the pigeon pea
should be rotary chopped before sowing and the regrowth sprayed

40 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 41


BOX 14. Critical points: Pigeon pea undersown in
maize for stubble grazing
(i) Pigeon pea seed may be mixed with the fertilizer and sown in
the maize row – it must be of high vigour and sown the same day
as mixed;
(ii) Second year plants, if not wanted, are killed by ground level
rotary chopping followed by a directed narrow band in-row
application of glyphosate or sulfosate on the re-sprouts (if any) –
this can be achieved by rotating fan nozzles to a diagonal position
and adjusting application rates accordingly, giving a savings of
over 50 percent on herbicide cost – application must follow the
original planter rows and cover the same width as the planter.

with a systemic desiccant herbicide or a split dose of paraquat (1.5 and
1.5 L/ha of commercial product). If left to grow in the second year, it
will need to be regularly grazed or it will become too woody for a
combine cutter-bar or easy manual eradication. However, it could then
still serve as a source of firewood or fodder for a small farmer.

Grazing stubble in the dry season
Nonavailability of fences and watering facilities can limit the use of
stover by cattle. Electric fences are cheap, with an investment cost in
Brazil of about US$ 80–100 per kilometre. The ratio of stubble area to
permanent pasture area is crucial to performance in this system. The
greater this is, the more positive the impact on farm total carrying
capacity (for more details, see Box 15).

Pasture grasses need to be selected according to the soil fertility level and
internal soil drainage, a problem mostly confined to wetlands. Table 8,
for the Cerrado region, shows the fertility requirements of the principal
pasture grasses used in the wet/dry Brazilian tropics. There is very little
information on which grasses are suited to integrated crop–livestock
rotations in the Amazon region, but this table would be a good starting
point, except for poorly-drained soils, where of the recommended

Vol. 5–2007 41
Fao_21_48.qxp 7-05-2004 15:30 Pagina 42


BOX 15. Critical points: Grazing stubble in the dry season

(i) Cattle compete with the ZT cropping system for the residue –
incomplete soil cover leads to greater weed infestation, but a 70
percent cover is enough to prevent erosion in most circumstances;
(ii) Soybean haulm is especially susceptible to untimely rains that
accelerate its decomposition;
(iii) Mineral salt with a protein additive is essential to effective rumen
throughput and hence total feed assimilation, and is provided in
roofed salt troughs, usually not removed during cropping –
movable types on skids are preferable;
(iv) Putting the cattle on the stubble also permits deferring grazing on
permanent pasture to later in the dry season, with some recovery
from residual soil moisture and late rains.

grasses, only Paspalum atratum and Brachiaria humidicola are suited (the
native Pará grass, Brachiaria mutica is also suited to wetlands), while
Andropogon gayanus and Setaria anceps are suited to drier climates.
In spite of the availability of a wide range of tested pasture legumes,
there has been negligible uptake of mixed grass-legume pastures in
Brazil, due to:
(i) real or imaginary difficulties in grazing management of these
(ii) excessive cost of seed, due to inefficient establishment of the
legume and high market prices because of low demand; and
(iii) lack of efficient establishment systems for mixed pastures, which
could start by undersowing legumes in maize and direct drilling
the grass at the onset of the following rains, when the legume
would have the advantage of an established root system and be
able to compete with the grass coming up from seed. Landers
(1994) indicated how to establish pasture legumes with maize
using alachlor as a selective herbicide for both. On-farm research
is required on these aspects.
For pasture grasses, Table 8 can be used either for initial species
selection, or to indicate when corrective P fertilization should be carried
out. Full fertilizer and lime recommendations for these forages are found

42 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 43



Andropogon gayanus cv. Planaltina Low 2 to 9
B. decumbens cvs. Basilisk, IRI 562 Low 2 to 9
B. humidícola cv. IRI 442 Low 2 to 9
B. ruziziensis Low 2 to 9
Paspalum atratum cv. Pojuca Low 2 to 9
B. brizantha cv. Marandu Moderate to high 3 to 15
Setaria anceps cv. Kazungula Moderate to high 3 to 15
P. maximum cvs. Vencedor and Centenario Moderate to high 3 to 15
P. maximum cvs. Colonião, Tanzânia-1, Very high 5 to 21
Mombaça and Tobiatã
Pennisetum purpureum cvs. Elefante, Napier Very high 5 to 21
Cynodon spp. cvs. Coastcross and Tifton Very high 5 to 21
Sugar cane Very high n.a.
* Mehlich (g/m3) Low value corresponds to <15% clay and high value to >60% clay i.e. lower P demand on soils
with lower C.E.C.
Source: Sousa and Lobato (2003), with addition of some cultivar names.

in Sousa and Lobato (2003). The traditional Molasses and Jaraguá grasses
(Melinis minutiflora and Hyparrhenia rufa), adapted to poorer soils, have
been out-performed by the Brachiarias and their use is now rare. Quinn
et al. (1963) showed the inferior performance of Molasses grass
compared to Colonial Guinea Grass in all situations, and inferior
performance of Jaraguá in high fertility situations, but holding its own
under low fertility.

Pearl Millet, Pennisetum americanum. This can be used purely for
biomass production, for high-quality grazing, for hay or for seed,
although rarely yielding more than 500 kg/ha of grain in late autumn
sowings, due to rainfall limitations; the protein content of its grain is
almost twice that of maize (Scalea, 1999). Harvest is by combine
harvester (by hand on small areas) and most farmers produce their own
seed. Breeding programmes have still not released hybrids; these should
make millet a good option as a grain crop for very late summer plantings
as it can extract nutrients from depth and can be sown on residual
fertility, with just a small top-dressing of N.

Vol. 5–2007 43
Fao_21_48.qxp 7-05-2004 15:30 Pagina 44


Pearl millet can be a host for army worm, but it controls nematodes
well and its residues apparently encourage less slugs in irrigated
conditions than those of maize. Millet is a slow starter, accelerating
development from about 30 days after sowing. After the first month, in
spring it can produce over 100 kg DM/day, but much less as days get
shorter. Millet can smother weeds sufficiently to eliminate, or reduce,
post-emergent herbicide use and has a similar effect in autumn once a
good cover is achieved. Under grazing it should not be allowed to initiate
flowering, in order to continue in vegetative mode.

Black Oat , Avena strigosa. This is not widespread in the tropics and is
limited to latitudes close to the tropic of Capricorn (São Paulo, Paraná
and Mato Grosso do Sul states) or altitudes above 1 000 m, as it does not
take well to heat. It dies back in summer and only requires rolling,
obviating pre-plant desiccation and often eliminating the need for post-
emergence herbicides, besides acting as a good recycler of nutrients and
incorporator of calcium in depth by translocation to its roots. When
grazed it should not be allowed to initiate flowering to maintain it in the
vegetative phase.

Forage Radish, Raphanus sativus. This crop has a deep tap root and is a
good recycler of N and other nutrients and has excellent forage value. It
is more a subtropical crop and performs best near the tropic of Cancer,
but has been used successfully at 16º S and 1 000 m altitude. Its small
seed requires care in sowing. This crop is a good precursor for maize.
Forage radish’s thick, carrot-like roots act as biological sub-soiling
agents. It may advantageously be mixed with other cover crops,
especially black oats, giving complementary cover, with a different
growth cycle. The Instituto Agronômico do Paraná has released
improved varieties of both black oats and forage radish.

Cut forage and silage crops
Commercial hybrids of maize and forage sorghum or sorghum x Sudan
Grass hybrids are the principal forages used for silage, and some open-
pollinated varieties from the Guinea coast of Africa have shown promise
(Séguy and Bouzinac, 2002). Resprouting ability after the dry season is

44 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 45


an important quality to evaluate in sorghums. Sugar cane, elephant grass
and recently, Brachiaria brizantha cv. Marandu are the most important
plants for cut forage. The importance of the latter is that it can fit into a
rotation as a catch crop, whereas sugar cane and elephant grass occupy
the same land for several years of ratooning. Kepler (2000) recommends
cutting maize silage no lower than 30 cm, leaving the potassium-rich,
indigestible lower stems. The recommended stage for cutting maize for
silage is full grain maturity (black layer at the base of the grain).
Sunflower can also be used for silage.

Pasture and grazing management
The merits of continuous versus rotational grazing will not be discussed; in
general, the former show superiority at lower stocking rates. Rotational
grazing is generally preferred in ICLZT systems in tropical Brazil. Table 9
gives a guide to managing the main grasses. On a pasture which is still in


GRASS ENTRY (cm) EXIT (cm) PERIOD (days) (days)
Andropogon 40–45 15–20 1–3 28–30
Brachiaria brizantha 30–40, 50–60 15–20, 30 1–3, 1–7 28–30, 28–42, 30–35
B. decumbens 30–40, 50–60 15–20, 30 1–3, 1–7 28–42, 30–35
B. ruziziensis 30–40 20–25 1–7 28–42, 30–35
P. maximum 50–60, 80–100 25–30, 40–45 1–3, 1–5, 7, 7–10 35, 35–40, 35–42
P. maximum cv. 50–60, 70–80 25–30, 35–40 1–3, 7 35–42
P. maximum cv. 50–60 25–30 7 35–42
Pennisetum purpureum 110–120 50–60 1–3, 3–7 35–45, 45
Schumach. cv. Napier
P. purpureum cv. 60–80 20–25, 30–40 1–3 45, 45–50
Elephant Grass
Hyparrhenia rufa 30, 30–40 10, 15–20 1–3, 7–10 30, 35–40
(Nees) Stapf
Digitaria decumbens 25–30 10–15 1–3 21–28
Setaria 30–40 15–20 1–3 30
Cynodon nlemfuensis
and Cynodon dactylon 25–30 5, 10–15 1–3, 1–7 21–28, 25–30

Melinis minutiflora
30–40 15–20 3–7 35 – 45

Source: Bonamigo (1999).

Vol. 5–2007 45
Fao_21_48.qxp 7-05-2004 15:30 Pagina 46


the maintenance phase, subdivision and rotational grazing on a 30–40 day
interval can give a significant boost to production. In this situation, the
investment in more animals to profit from the higher carrying capacity is
greater than the investment in electric fences (now common) and
watering facilities (Smith, 2004). See the example of Case Studies No. 1
and 6 in Chapter 5. Continuously grazed permanent pasture, or
temporary winter pasture, both need to recover after the first rains; the
latter to regenerate biomass for desiccation in order to provide adequate
residue cover (at least 30 days). Temporary grazing on the crop area must
allow grazing on part of the permanent pasture to be deferred until the
onset of the rains, preferably in bottomland with better moisture
availability (see Case Study 1 in Chapter 5). At this critical period beef
prices are still high and stockyard or supplemental feeding with silage or
green chop may be attractive alternatives (See Case Studies 1, 2 and 5).
Kluthcouski et al., (2003) recommend an initial light grazing of
grasses undersown in maize directly after harvest, to encourage tillering
and to correct any etoliation problems. For mature pasture, these
authors recommend grazing at the highest level of protein (flower
initiation), however, in practice, as flowering is not year-round, the
critical height in Table 9 can substitute for this criterion.
Weight gains of 1 kg/day were recorded between April and June on a
sorghum + B. brizantha mix, stocked at 3.5 head/ha. This pasture also
furnished maintenance grazing until the onset of the rains (Kluthcouski
et al., 2003). “Early bite” with forage sorghum intersown with a
permanent pasture species gives grazing of the sorghum at 30–45 days
after sowing. A comparison between the calendars of ZT and
conventional tillage in crop–pasture rotations, using the above system,
was made by Dos Santos (2003), as shown in Figure 8. The elimination
of time for land preparation and the inclusion of forage sorghum for
early bite gave a considerable advantage in pasture carrying capacity
over the time of the comparison.
Adjustment of stocking to pasture production is necessary to avoid
overgrazing and reduction of regeneration capacity. Martha Junior et al.
(2004) have developed a methodology to calculate the grazing and rest
periods in a rotational grazing system according to desired weight gain,
forage availability and pasture recuperation capacity, based on the values

46 Integrated Crop Management
Fao_21_48.qxp 7-05-2004 15:30 Pagina 47




Animals graze the new pasture 30 days after sowing



Animals graze the new pasture 60 days after sowing

Liming broadcast 1 st Heavy discing Sowing

Grazing Liming Followed New pasture
by 2 nd discing high stocking

Desiccation and Levelling disc New pasture
sowing low stocking

FIGURE 8: Comparison between the calendars of a ZT-based crop–pasture rotation and one based
on conventional tillage
Source: Dos Santos (2003).


Return period (days) 56 42 38 36
Forage Production (kg DM/day) 30 50 70 120
Grazing efficiency (%) 45 50 55 55
200 kg 1.18 2.18 3.36 5.75
300 kg 1.30 2.42 3.72 6.38
400 kg 1.41 2.61 4.03 6.90
Source: Martha Junior et al., (2004 ).

Vol. 5–2007 47
Fao_21_48.qxp 7-05-2004 15:30 Pagina 48


in Table 10. The net offer of forage in a paddock, adjusted for grazing
efficiency, is balanced by the total demand for a given number of animals
and the number of grazing days in the rotation. This could be refined by
using total digestible nutrients (TDN).

The establishment and management of grass-legume pastures have, so
far, been beyond the management capacity of most farmers in tropical
Brazil, although adapted cultivars and experimental results are
available. Where it has been done, results are highly encouraging, see
below, but management for legume survival is critical. Undersowing,
oversowing, sowing the legume before the grass and seed pelleting or
coating require more study.
Pigeon pea is a notable exception and its use in ZT sowing into
pasture (or in maize, for stubble grazing — Von Schaffhausen, 1949), is
well-established and increasing. The most successful perennial legume
has been Stylosanthes guianensis cv. Mineirão, but it is slow to establish
and its seed is high-priced (normal for a new cultivar with little outlet).
Liveweight gains on different pastures were maximized by using a
Stylosanthes guianensis protein bank, with Brachiaria brizantha, giving
gains of 577 kg/ha/year as compared to 258 kg/ha/year with B. brizantha
alone. Another grass/legume pasture (B. decumbens + S. guianensis) gave
an intermediate performance (Macedo, 1999). Leucaena leucocephala
protein banks suffer from severe dry season leaf drop, making them
only suitable to very restricted areas with a relatively high water table,
but which do not become waterlogged in the rains. Protein banks avoid
the complications of managing a mixed sward, and access once a week
can be adequate.

48 Integrated Crop Management
Fao_49_60.qxp 24-04-2004 13:08 Pagina 49

Mechanized operations
in zero tillage and soil
fertility management

Operations are described in order of occurrence, starting from managing
residues, through pre-plant desiccation, planting-drilling and post-
planting operations. This is followed by considerations of liming, gypsum
application, summaries for N, P and K and a section on soil sampling
under ZT. Liming and fertilizer application use conventional technology
and are therefore not discussed in detail.

The first rule in residue management is to ensure even distribution of
crop residues after harvest; this helps to suppress weeds and facilitate
planting/drilling operations. Combine harvesters should use a straw
spreader or chopper to ensure this. Argentine rolls (knife rollers) are
effective on some cover crops, especially black oats and should cross the
direction of future planting rows. Vertical flail choppers and horizontal
rotary choppers may be used to flatten maize or cotton stalks and break
down clumps in pastures prior to desiccation. The latter tends to produce
windrows, and planting diagonally across windrows reduces clogging.

Pre-sowing desiccation is the most important operation in ZT. Weeds not
properly controlled during desiccation are very costly to eliminate with
selective herbicides. After sowing, the same selective herbicides are used as
for conventional tillage; local or regional research-based information should
be sought, with recommendations for the specific conditions of the farm.

Vol. 5–2007 49
Fao_49_60.qxp 24-04-2004 13:08 Pagina 50


Marking for desiccant application is absolutely necessary; besides
surveying the fields and leaving permanent markers in fence lines,
marking foam applicators at the tips of the spray boom are available
and Brazilian farmers have developed a system (shown in Figure 9),
which uses two tractors and a cable attached to the rear of the first
tractor and the front end of the second. Since there is more weight on
the rear wheels of the first tractor, it keeps the second in line. The first
tramline is marked with poles at half the sprayer width from the field
edge and as the first tractor goes along this
line, the second leaves tracks in the stubble;
the first tractor then follows the second
tractor’s tracks successively and the double
passage leaves enough of a track to be
followed during the spraying operation.
Planting lines should be parallel to the
tramlines (which help to keep tractor
drivers straight), or may be transverse,
accepting a minor loss of stand in the FIGURE 9: System of marking for
pre-plant desiccation and later
tramlines. Tramline spacing should be a spraying operations, using two
tractors and a cable for a gauge.
multiple of planter row width.
Sprayers range from 20-litre backpack, single-nozzle types to self-
propelled versions with 24-metre booms and 2 000-litre tanks. Spray
planes are widely used by large farmers.2 Spray technology is a
specialized affair and will not be dealt with in depth. Systemic
desiccant herbicides must be applied as large droplets; low volume
applications improve herbicide absorption and sprayer efficiency, but
require higher skill in application. Reducing the amount of water
applied from 200 litres/ha to 30 litres/ha substantially increases area
sprayed daily and can permit reduction in chemical application rates.
Stressed plants do not absorb herbicide effectively; the best timing is
during vigorous growth, avoiding the middle of the day with its high
temperature and low humidity.
An innovation for large farms is the drag system of applying
chemicals. Two tractors drag a line of nozzles fed from a hose attached to
Although application of desiccants (herbicides) by airplane is often done in practice, FAO does not
recommend this technique, due to the environmental considerations involved.

50 Integrated Crop Management
Fao_49_60.qxp 24-04-2004 13:08 Pagina 51


a cable between them – the length of the cable can be up to 100 metres,
giving a very high work rate for desiccant application.3
Desiccant application rates in pasture depend on the species involved.
Brachiaria ruziziensis requires about half the dosage of glyphosate or
sulfosate/ha compared to B. brizantha; B. decumbens is intermediate,
while B. humidicola, panicums and other grasses require even higher
dosages – a summary of manufacturers’ recommendations for tropical
Brazil is shown in Table 11.


B. ruziziensis B. brizantha , 2
P. maximum 2,3

B. decumbens 2
cv. Mombaça
P. maximum 2
Cynodons 2,3

(except cv. Mombaça) P. notatum 2,3

A. gayanus 2,3

B. humidícola 2,3

1–1.5 kg/ha 4
2–2.5 kg/ha 4
2.5–3.0 kg/ha 4

Wettable granules (commercial product) assumes 100 L/ha application rate of the spray mixture.
For medium and low susceptibility grasses, apply 2/3 of the dose and wait 15–20 days before applying the rest.
For the low susceptibility grasses apply 2/3 of the application in autumn and the rest in spring.
For plants germinated in the same year, or still with a small root mass, rates can be reduced.

Source: Monsanto Brazil.

Choice of herbicide depends on a survey of the weed spectrum and
stage of growth, leading to an evaluation of susceptibility. Selective
herbicides may be mixed with the desiccant for specific weeds – check the
local-regional weed control recommendations and a chart of
compatibility for active ingredients and spreaders.
Spot-spraying, which can be very cost-effective to control isolated
patches of problem weeds, should be done before the overall
desiccation, or later as a complement to eliminate misses. This permits
higher application rates or admixtures of specific herbicides for
problem weeds. It can be done manually with backpack sprayers or
with lances coupled to the tank of a boom sprayer (isolating, or
removing, the booms), or even by hand-hoeing, or pulling, if
This drag system, although convenient for many farmers, is not technically recommended by FAO as a
controlled quality application.

Vol. 5–2007 51
Fao_49_60.qxp 24-04-2004 13:08 Pagina 52


populations are small. In the case of a heavy green mass at desiccation
(cover crop or weeds), up to three weeks should be allowed between
spraying and sowing, to allow dissipation of allelopathic products from
root decomposition (Constantin and De Oliveira Jr., 2005) which
depress yields as they decompose, as shown in Table 12. For further
information about spraying desiccants, see Box 16.


(Kg/ha) (Kg/ha)

Maize 1 110 654
Soybeans 468 408
Local farmer average
674 Not determined
yield with ZT soybeans
Source: Constantin and De Oliveira Jr. (2005).

BOX 16. Critical points: Spraying desiccants and other

(i) Water must be clean and free of suspensions – clay and organic
matter absorb the spray chemicals and reduce effectiveness;
(ii) Fine sand or other particles can clog nozzles;
(iii) Always test sprayers before use, for correct operating pressure,
nozzle overlap, individual discharge (volume/time) and replace
worn nozzles etc; calibrate for application volume;
(iv) To judge application rate of chemical required, imagine the root
reserves of the plant; the greater the live root mass, the higher the
application rate;
(v) Wait for dew to dry before spraying, or leaf runoff may reduce
active ingredient absorbed;
(vi) Do not sow before desiccation, as rain may not allow desiccant
application before crop emergence;
(vii) Recently germinated plants, with small root mass, are easily
controlled with low dosages of contact herbicides.

52 Integrated Crop Management
Fao_49_60.qxp 24-04-2004 13:08 Pagina 53



These operations are normally carried out on
desiccated areas, with or without cover
crops, or on rolled cover crops, such as black
oats, which dispense with desiccation.

Manual jab planters have been adapted for
ZT by reinforcing and narrowing the tips to
give better penetration into untilled soil. They
are commonly equipped with a double tip and
apply fertilizer at the same time as the seed.
PLATE 12: Hand jab planter,
Both fertilizer and seed flows are adjustable. commonly called a “matraca”, with
Plate 12 shows a hand jab planter; to operate it both fertilizer and seed hoppers.
is jabbed into the ground with the point closed and then the handles are
pulled apart, allowing seed and fertilizer to drop into the planting hole.
On closing, the seed and fertilizer points are re-charged. A normal work
rate for maize planting in 1m rows would be 1 ha/man day.

Animal drawn planters. There are a variety of models in Brazil, mostly
single row types, pulled by a single horse or mule or a pair of oxen (see
Plate 13). Most have a trash-cutter disc up front, followed by a fertilizer
knife and an eccentric double disc for the seed (see Plate 14). The simplest
model with no trash disc and no furrow closing wheels is not suitable for
heavy residues and can leave seed uncovered. Some versions have depth
wheels for both seed and fertilizer, others only one for seed; their
position and size are variable and furrow-closing mechanisms are also
varied, sometimes absent, the most common being double-angled wheels.

PLATE 13: An animal drawn drill under PLATE 14: Details of construction of another
demonstration, planting into desiccated pasture. model.

Vol. 5–2007 53
Fao_49_60.qxp 24-04-2004 13:08 Pagina 54


Fertilizer distribution is of variable quality between models and seed
distribution is invariably with a horizontal perforated disc. Some models
use discs commonly available from large manufacturers, which facilitates
local availability when different spacings are required for new crops. A
canola or sorghum disc works well with pearl and finger millets.

Tractor-mounted and tractor-drawn planters and drills. Tractor-
mounted models are limited to small units (maximum five rows) as well
as self-propelled walking tractor machines (see Plate 15), some with a
rotary hoe furrow opener. Larger units are tractor-drawn with their own
transport wheels and vary from 4 to 20 row units – they are usually
Three basic types of seed distribution are available:
• planters using perforated disc or pneumatic seeders;
• small-grain drills using continuous flow distribution of seed,
mechanical or pneumatic;
• multi-purpose machines for all grain sizes.


PLATE 15: A self-propelled single axle walking tractor, with attached two-row direct-planter,
suitable for CA systems.

54 Integrated Crop Management
Fao_49_60.qxp 24-04-2004 13:08 Pagina 55


In tropical Brazil, the first category is overwhelmingly the most
important. Since small grain production is minor compared to maize,
soybeans and cotton, seed drills are restricted to rice, wheat and barley;
extra investment in a drill is considered unprofitable for cover crops in the
tropics. This is a short-term view, since the common practice of
incorporating broadcast cover crop seed with a light discing incurs some
loss of surface residues, which are difficult to maintain anyway, and which
also wastes precious moisture, through evaporation and delayed planting.
Trash-cutting discs, which have been getting larger and vary from
40–56 cm in diameter, may be straight or fluted, (wavy) or rippled to
ensure proper turning through good contact with the soil, with the fluted
types more suited to sandy soils. The straight type is the most common
in Brazil. Fertilizer coulters have three basic variants:
• the guillotine knife coulter closely coupled to the trash-cutting disc
with a small vee-shaped guillotine attached to the knife, acting as a
disc cleaner (the very small clearance is critical to performance and
the size of the guillotine is smaller in heavier soils);
• a knife coulter some distance behind the trash-cutting disc
following in the same furrow as the trash disc;
• offset double discs with different diameters receiving fertilizer in the
gap between the discs.
Offset double discs should only be used when soils are well-structured
(after 4–5 years of continuous ZT) and usually not for maize, cotton or
sowing in desiccated pastures (see Plate 16). Since it is difficult to offset the
seed disc from the fertilizer coulter, many farmers weld a small piece of
chain, or wings, behind the coulter to improve the mix of fertilizer with
soil and reduce the potential for seed burn from high K or N fertilizers.
This problem can also be addressed by
broadcasting most of these fertilizers after
sowing, with potential gains in efficiency from
reduced leaching and less down time for the
planter. Normal fertilizer placement aims at a
depth of 12–15 cm, usually directly below the PLATE 16: A seed unit showing the
seed. For irrigated phaseolus beans it has been double eccentric seed disc,
adjustable depth control and
advantageous to split the depth of fertilizer furrow closing wheels – note that
the units are offset to avoid
application, placing half at a shallower depth. clogging with surface residues.

Vol. 5–2007 55
Fao_49_60.qxp 24-04-2004 13:08 Pagina 56


Seed distribution mechanisms for planters are still dominated by the
perforated disc, but pneumatic seeders are available at a higher price.
Discs with a double row of perforations halve the disc rotation speed
and improve distribution, allowing faster planter operation. A practical
limit for planting speed is 6 km/hour, even for soybeans and pneumatic
seeders (CAT Uberlândia, 2005). Uniformity of seed and fertilizer
distribution is variable between planter-drill models and has become
more crucial as different cultivars of maize and soybeans require
different populations for top yields. Air seeders are not in general use in
tropical Brazil and small grains are sown with the traditional drill
mechanism or with multi-grain-size universal machines, most common
on small farms in South Brazil.
Recent modifications have been: Increased trash disc diameters,
pantographic suspensions for individual rows, independently hinged
sections of the planter to allow better seed depth uniformity when
planting in straight lines over the contour banks, folding planters for
transport and easier manual adjustments. All tractor models use seed
depth control wheels; the closer these are to the point where the seed
enters the furrow, the better will be the uniformity of seed emergence.
The principal types of furrow-closing wheels are:
(i) Paired slanting wheels whose angle and pressure are adjustable;
(ii) Single over-the-top rubber-covered press wheels; or
(iii) A single slanted metal press wheel.
The use of paired opposite discs for furrow closing has been
discontinued, but some farmers have replaced the wheels with a section
of looped chain. Farmers still make their own modifications to facilitate
field adjustment; many of these have been adopted by manufacturers, e.g.
the guillotine disc cleaner.
On soils of over 40 percent clay, planter adjustment is especially
tricky. It requires some persistence to obtain satisfactory performance,
but this has been achieved on soils varying from 5–60 percent clay.
In Plate 17 is a farmer modification to plant pasture seed after
soybean harvest or coincidently with maize, using a pneumatic
distributor to place the grass seed between the double seed discs – with
the consequent in-row soil movement, most seed is covered by less than
2 cm of soil.

56 Integrated Crop Management
Fao_49_60.qxp 24-04-2004 13:08 Pagina 57


PLATE 17: Farmer modification to plant grass seed after soybean harvest (or for interplanting in
young maize). The white boxes behind the planter are the pneumatic unit seed distributors – the
delivery tube is inserted between the double seed discs.

Another farmer modification can be seen

in Plate 18. In this case, an innovative farmer
has attached a thick wooden pole to be
dragged behind his planter. This modification
is very helpful to assist with covering the
seeds and redistributing the crop residue
more evenly.
PLATE 18: Another farmer
In Box 17 some specific points are given modification to improve seed and
residue cover.
about using planting and drilling machines.

BOX 17. Critical points: Planting and drilling
(i) Planters and drills must be levelled along the planting line before
work, so that fertilizer and seed can be adjusted to the desired depths;
(ii) “Hair-pinning” of residue below the trash disc often occurs with
wet or wilted residues, or on very wet soils, and then planting must
be suspended;
(iii) Always use a fertilizer boot, guillotine or narrow tine, for maize or
cotton or for planting into desiccated pastures;
(iv) Check clearance of guillotine from trash disc according to
manufacturer’s specifications;
(v) Check for irregularities and sharpness of discs to prevent clogging;
(vi) Post-plant operations in the row must use the same number of
rows as the planter and follow in its tracks.

Vol. 5–2007 57
Fao_49_60.qxp 24-04-2004 13:08 Pagina 58


Liming. The effectiveness of broadcast lime application without
incorporation has been long proven in ZT, as the lime enters the soil
through old root holes and fauna-made galleries, more slowly through
leaching of calcium and magnesium, and rapidly through translocation in
the roots of black oats (Lopes, 2004; Bernardi et al., 2003; Sá, 1993).
Maintenance lime requirements are greatly reduced under zero, compared
to conventional tillage, due to less anaerobic conditions for organic matter
breakdown (Lopes, 2004). Farmers report going over 10 years without
liming after adopting ZT, also due to earlier over-liming or large diameter
lime particles taking longer to react. Care must be taken not to apply more
than 0.5–1 tonne/ha (depending on soil texture and CEC), otherwise there
can be an over-reaction at the soil surface, raising pH to above 7, when
micronutrients, especially manganese, become less available. Manganese
deficiency in soybeans shows as interveinal chlorosis and can easily be
remedied by spraying nutrient solution containing Mn.
Excessive liming may also be responsible for cementation problems on
silty sands (P. L. de Freitas, personal communication, 2005). Critical base
saturation percentage (V percent) under ZT is 40; when it falls below this,
apply lime to raise it to 50 percent (Sousa and Lobato, 2003). Sousa and
Lobato (2003) also indicate that lime demand under ZT is reduced by
35–50 percent compared to conventional tillage; local recommendations
for lime application often refer exclusively to the conventional situation
and thus over-estimate the demand in ZT. There is increasing evidence
that low base saturation levels are less limiting in soils with many years
of ZT and consequently higher SOM and residue levels (L. Séguy,
personal communication, 2004).
Planning the conversion of pasture to crops allows broadcast lime
application one rainy season before planting the crop, ensuring longer reaction
time and less likelihood of limitation to crop yield. In extreme cases, in-row
applications of very finely ground lime (“filler”) of 150–300 kg/ha can be
made in the crop row, with a third hopper (Séguy et al., 1998), where liming
may not have been adequate; local calibration of this rate would be necessary.

Gypsum. This is applied broadcast in small quantities to supply soluble
calcium to penetrate to the subsoil and correct deficiencies of that nutrient,

58 Integrated Crop Management
Fao_49_60.qxp 24-04-2004 13:08 Pagina 59


or to reduce aluminium toxicity in the subsoil, or simply to correct sulphur
deficiency throughout the root zone. According to Table 13, derived for
latosols (Ferralsols), application levels rarely exceed 600 kg/ha.


Low 10 x Clay %
Medium 5 x Clay %
Source: Sousa and Lobato (2003).

Higher application rates may be needed for in-depth correction of
Ca deficiency and neutralization of subsoil Al. Gypsum may be
applied with lime.

Phosphorus nutrition. Phosphate should not be broadcast together with
surface application of lime in ZT since this causes reduced P availability
due to a surface pH close to 7.0 (Sousa and Lobato, 2003).
Higher clay contents require increased P fertilization since a higher
application rate is needed to shift the available P content.
Surface application of phosphatic fertilizer can only be made when
the soil’s P status is adequate to supply the needs of the present crop, the
applied fertilizer being to maintain the system. Short cycle crops are
more demanding in P than longer cycle ones, which have more time to
extract it. Pasture grasses and legumes are more effective in extracting P
from the soil than field crops: over seven years continuous cropping on
a tropical latosol (Ferralsol), soybeans extracted 21 kg of P2O5/ha,
while two years of soybeans followed by five of B. humidicola extracted
50 kg of P2O5/ha (Vilela et al., 2004).

Nitrogen nutrition. High organic matter and crop residue levels under ZT
mean that more of the N in the system is sequestered. Therefore, higher
N fertilizer applications are needed in the first years until an equilibrium
is reached (Sá, 1993). For maize, an extra 25–30 percent of N is required;
this also applies to other non-leguminous crops and phaseolus beans.

Vol. 5–2007 59
Fao_49_60.qxp 24-04-2004 13:08 Pagina 60


There is evidence that the enhanced biological activity of CA systems
requires earlier N application than conventional tillage; in some soils this
can be 100 percent pre-plant, but in years of exceptionally heavy rains this
may not be an advantage (Lopes, 2004). Many farmers consider the
reduction in the weight of fertilizer carried in the planter as a sufficient
operational gain to justify pre-plant application of N. Local research is
required on this point.
Urea should always be incorporated and not top-dressed, since crop
residues liberate urease which breaks urea down rapidly and N is lost by
volatilization as ammonia – in extreme cases (high temperatures and no
rain) such losses may approach 70 percent (Lara Cabezas, 1998).

Potassium nutrition. There is a build-up of K in the top 10 cm of soil
under ZT and little evidence of leaching. Since K fertilizers cause seed
burn and it is not usually possible to place fertilizer to the side of the seed
in ZT, half, or all, of the K fertilizer can be applied broadcast. Split
applications may improve efficiency. Alternatively, the fertilizer boot can
be modified to apply the fertilizer at two depths; this has been effective
in heavily fertilized phaseolus beans under centre pivot irrigation.

Soil sampling. Brazilian practice in ZT is to sample at 0–10 and 10–20 cm,
to take account of the high nutrient concentration at the soil surface, due
to nutrient recycling through surface residues. Further sampling is
advisable at 20–40 cm to determine if there is a calcium deficiency or toxic
aluminium levels which will impede root development. In ZT, soil
sampling needs to span the whole row width, with a specialized broad, flat
(vertical) sampler, otherwise fertilizer concentrations in the row will
distort results, (unless a very high number of random samples are taken).

60 Integrated Crop Management
Fao_61_74.qxp 24-04-2004 13:09 Pagina 61

Technical and financial
analysis of integrated
zero tillage rotations

This chapter uses six case studies to illustrate the technical and financial
aspects of ICLZT, compared to the situation prior to adoption of the
technology. Results are conservative because the farms were all practising
CA before adopting ICLZT and had an above-average management
level, so the differences introduced by ICLZT were less marked than if
the models had started with average management, using conventional
tillage. Integrated crop–livestock zero tillage technology is new in the
tropics (Broch et al., 1997) and no farms were encountered which had
gone straight from conventional tillage to ICLZT without an
intermediate stage of CA crops without pastures. The first case study is
described in some detail to show the steps in the farm system’s evolution
(see Box 18), while case studies 2–5 are summarized (Landers et al.,
2005). The sixth case study was included to illustrate its advanced
management with a one-year ley.

General description. The farm selected belongs to a traditional farmer in
Vianópolis, Goiás state, who has modernized his operation from 1979,
when he purchased a run-down cattle farm with degraded B. decumbens
pastures, typical of the lower half of the 60 million ha of pastures today
in the Cerrado region; he adopted a full ICLZT rotation in 1997.

Vol. 5–2007 61
Fao_61_74.qxp 24-04-2004 13:09 Pagina 62


The farm has 1 175 ha, including mandatory reserves of 280 ha (24 percent
of total area), leaving 600 ha of crops (190 ha irrigated and 410 ha in
rotation with pasture), 285 ha of pasture and 10 ha of other uses. Annual
rainfall is 1 638 mm, latitude 16º 45’ S, longitude 48º 30’ W.

BOX 18. Phases of development in the adoption of
CA with ZT
1979 Removal of erosion gullies with a crawler tractor for the large
gully and deep ploughing for small ones and rills, followed by
planting of B. decumbens, undersown in rice using limited fertilizer,
incorporating 3 tonnes/ha of lime and re-making of contour
banks for soil conservation.
1980 Fencing off of the river-edge gallery forest to allow self-
1982 Production of 1 100 tonnes/yr of silage for dry-season supplement;
1980 - 1987 Conventional tillage with rainfed soybeans and maize, separate
from the pasture area; pasture renovation with rice and undersowing
of Brachiaria brizantha, initial pasture subdivision into 50 ha paddocks.
1987-1997 Gradual adoption of ZT.
1995 Introduction of improved pasture management with pigeon pea
planted directly into pastures and installation of watering
facilities for a beef fattening operation on grass;
1995 Installation of centre pivot irrigation on 190 ha;
1996 Introduction of an integrated crop–livestock rotation, with
subdivision of pastures using electric fences into 1 and 2 ha units;
rotational grazing on a 42-day basis.
2000 Conversion to a cow-calf (weaner) operation, with supplemental
silage limited to 400 tonnes/yr for supplemental (creep) feeding
of calves before weaning and for emergency drought reserve.

The investment model. This model represents the evolution of Cerrado
biome farming from medium to high level technology in a mixed
crop–pasture farm, which was originally solely pasture. The absentee
owner lives 120 km away. The high management level is obtained
through local professional consultants (veterinarian and agronomist).
This is now common in the region.

62 Integrated Crop Management
Fao_61_74.qxp 24-04-2004 13:09 Pagina 63


The 20-year investment model considers the situation before the
farmer decided to improve his management with ICLZT in 1995 and
compares this with the results of ICLZT in 2004. Crop yields used for
the “without project” situation correspond to the present day
expectancy under the “without project” management and incorporate
genetic yield improvements over the time-span between the “before”
and “with” project situations on this farm. The use of historic yield
data of ten years ago would have exaggerated the gains with ICLZT
since there have been considerable gains in genetic potential of both
soybeans and maize since then. Zero tillage adoption preceded the
ICLZT technology on this farm, thus the comparison indicates the
gains due to ICLZT alone, excluding the effect of conversion from
conventional tillage to ZT. The “without project” situation, Scenario
(i), represents approximately the situation of the middle third of
producers in the Cerrado region and the “with project” Scenario (ii)
situation, that of the upper ten percent. Since the farmer had changed
his type of operation from a fattening to a breeding herd after adopting
ICLZT, the equivalent herd parameters for a cow-calf operation under
the “without project” management level, estimated by the farm
veterinarian, were used in the comparison. The lack of good quality dry
season feed in the “without project” situation affected calf weight at
weaning; ICLZT allows earlier sales at the same weight, but no price
increase was imputed in terms of quality, which would probably occur,
also due to upgrading with pedigree bulls. Higher cow and calf
mortality and lower calving and replacement indices occur in the
“without project” situation.
Both scenarios are placed in the same time frame, with February 2004
farm gate prices – the price for irrigated maize was increased by 10 percent
for out-of-season sale. In this model both rainfed and irrigated crops are
assumed to be 100 percent under ZT in scenario (i), with the exception of
conventional cultivation for pasture renovation. For purposes of
comparison, the same cropping patterns were used for both scenarios;
equipment only differed in heavy levelling disc harrow and plough for
pasture renovation in scenario (i). In scenario (ii) a ZT planter with
chisel-type furrow opener (Plates 26 and 27) for planting pigeon pea in
pasture and electric fences for pasture subdivision were used.

Vol. 5–2007 63
Fao_61_74.qxp 24-04-2004 13:09 Pagina 64


Without project
A medium technology level was used, with the pastures for the cow-calf
operation separate from the annual crops. Zero tillage was already used
in both rainfed and irrigated areas, with pasture renovation using
conventional tillage: the rainfed crop rotation was 3 yrs soybeans – 1 yr
maize. The farmer had intensified part of the crop area (190 ha), using
pivot irrigation. Irrigation is practiced on less than 5 percent of the crop
area in the region. Pasture renovation was done every six years, using
conventional land preparation and defraying the cost of establishment by
undersowing B. brizantha in a crop of upland rice. Paddocks were about
50 ha of B. decumbens and more recently B. brizantha. Winter carrying
capacity of pastures was low and offset by production of 1 100 tonnes of
silage to feed the whole herd from August to October. The farm
veterinary consultant’s estimate of 220 cows plus followers was used for
the “without situation” number of cows stocked, based on historical
information. The pasture area was permanent and separate from the crops.

The farmer directed his attention to improving the winter pasture
situation for the cattle as the best route to improve profit. In 1996, he
adopted a high-technology ICLZT rotation – annual rainfed ZT crops in
rotation with pastures and irrigated ZT cropping separate. This was
accompanied by more intensive grazing management through pasture
subdivision with electric fences.
The original 220-cow herd was increased gradually by retaining
replacement heifers until full carrying capacity under ICLZT was reached.
Herd size used with ICLZT (380 cows plus followers) was actual, from
farm records, and equal to the farmer’s estimate of optimum pasture
carrying capacity. Investment in higher pasture technology was stimulated
by the adoption of ICLZT and improvements in performance were
assisted by buying high quality bulls from progeny-tested sires.
Pasture stocking levels are considerably enhanced with ICLZT by:
(a) rotating with a crop phase;
(b) sowing pigeon pea into the pasture, with phosphate fertilizer
and a band application of desiccant herbicide to check the
pasture in the row;

64 Integrated Crop Management
Fao_61_74.qxp 24-04-2004 13:09 Pagina 65


(c) subdividing paddocks with electric fences; and
(d) oversowing millet into rainfed crops for winter stubble grazing.
The higher organic matter and the improved weed, pest and disease
control from the pasture phase contribute to 10 percent higher yields for
rainfed soybeans and maize, with the project. The crop budget-yield and
costs of pasture maintenance were considered equal after planting for
both scenarios. This is conservative, since the rotation of crops with
pasture will also reduce the costs of weed, disease and pest control;
records did not permit an accurate assessment of these aspects.
The pasture is renovated every four years, accompanying the crop
rotation. Summer pasture is subdivided with electric fences into 2 ha
paddocks, grazed intensively for seven days, with 35 days for recovery.
All pasture in rotation with crops is planted with high digestibility
Panicum maximum cultivars (Tanzânia and Mombaça), or with
Brachiaria brizantha. Paspalum atratum cv. Pojuca, which is less
palatable but tolerates poor drainage, is planted as permanent pasture in
the bottomlands. Renovated pasture is planted with pigeon pea, which
lasts about three years and acts as a winter protein supplement, increasing
average protein content in the diet and maintaining good rumen
throughput and hence weight gain. Three quarters of the soybean crop
area is oversown broadcast with pearl millet (Pennisetum americanum cv.
BN2) as the soybean is maturing. The cattle go onto the millet in soybean
stover in May, and in June-July-August onto the maize stover. In late
August, the whole herd goes back onto the upland P. maximum pasture
sown with pigeon pea. With the first rains (mid-Oct.), the cattle are
moved to the bottomlands, which give strategic grazing at this time,
allowing the upland pastures to recover. About 30 ha of soybean stover
sown to millet is left for seed and grazed after harvest. Fallen millet seed
and weeds germinate with the first rains (Sept–Oct) and produce ground
cover, which is desiccated in Nov–Dec for the next crop.

Irrigated crop management – with and without project
With and without ICLZT, the three centre pivots, covering 190 ha,
receive identical management. Cropping intensity is 200 percent,
preferred over higher intensity because of easier planting-harvest
schedules. The rotation is: year 1 maize-for-grain/kidney beans; year 2

Vol. 5–2007 65
Fao_61_74.qxp 24-04-2004 13:09 Pagina 66


soybeans/wheat; year 3 maize-for-seed/kidney beans. Average irrigation
period for beans is 90 days per crop; maize-for-grain receives an average
of 35 days of supplemental irrigation between September and October;
wheat consumes some 400 mm/crop in 100 days and soybeans are not
irrigated. Irrigation is guided by tensiometers. Irrigated stovers are not
grazed. Yields are equal with and without project, as this area is kept
separate from the cattle/rainfed crops rotation.

Analysis of the Model Results
Herd performance improved after adoption of ICLZT and rotational
grazing, as shown in the summary in Table 14. The number of cows
increased from 220 to a steady 380 from year 9 onwards and stocking rate
on the pasture increased from 1.0 to 1.76 AU/ha. Maize silage area was
reduced from 22 ha to 8 ha with ICLZT, where it is only used as
supplemental feed for unweaned stock and as a tactical drought
reserve. Together with small increments in rainfed crop yields under
ICLZT (10 percent) due to rotation with pasture, this gave IRR and NPV
values as shown in Table 14.
For a modest incremental investment, a ZT planter and drill, electric
fences, establishment of direct-seeded pigeon pea in pasture and watering
facilities, IRR increases from 14.16 percent to 19.96 percent and average
annual net profit (undiscounted) from R$229/ha to R$372/ha, a gain of
63 percent, making this investment in technology extremely attractive to
the farmer who has the management capacity. If compared with average
management levels in the region, this increase would be even higher.
Table 15 shows the results of three alternative scenarios applied to
Case Study 1, compared with the “with project” situation, as Scenario
(i). In this scenario, the rainfed crop enterprise shows a small
deforestation mitigation potential (DMP) of 0.11 ha of native vegetation
preserved for each hectare under crops (due to small increments in maize
and soybean yields with project). In addition, the cattle enterprise shows
a gain of 0.76 ha/ha in total AU carrying capacity, giving a total
mitigation potential (crops + pasture) of 0.87 ha/ha in ICLZT.

Defined as the percentage capacity of expansion in production achievable by land use intensification within
the existing area.

66 Integrated Crop Management
Fao_61_74.qxp 24-04-2004 13:09 Pagina 67



Investment Yr 1 (R$ total) 1 290 363 1 268 581 1.72
New Investment Yrs 1–5 (R$/ha) 1 458 1 433 1.72
IRR (%) 19.96% 14.16% 40.98
NPV (R$ ´000) 1 454 183 542 340 168.13
Annual output of grain
(average) tonnes/ha
4.53 4.29 5.78
Net profit/ha/yr 372 229 62.64
Total area (ha) 1 175 1 175 0.00
Area in reserves (ha) 280 280 0.00
Area of rainfed crops yr 20 (ha) 410 410 0.00
Area of irrigated crops yr 20 (ha) 190 190 0.00
Area of pasture yr 20 (ha) 285 285 0.00
Percentage of pasture
(exploited area) 32.20% 32.20% 0.00
No. of cows yr 0 220 220 0.00
No. of cows yr 20 380 220 72.73
Total Animal Units yr 20 501 299 67.59
AU/ha yr 20 1.76 1.05 67.59
Source: Expanded from Landers and Weiss (2005).

Scenarios (ii), (iii) and (iv) begin with the typical degraded pasture
situation of a winter stocking rate at 0.5 AU/ha on the whole farm. In
Scenario (ii) crop expansion of 600 ha generates 0.68 ha/ha, plus 0.11
ha/ha yield gain and 0.13 ha/ha for a 13% gain in carrying capacity,
totalling 0.92ha/ha DMP.
Scenario (iii) shows renovating the pasture with lime and fertilizer,
with no crops and in scenario (iv) there is a pasture/crop ratio of 1:1,
with pasture renovation based on the residual fertility of the crop phase.
Both show greater mitigation potential (2.52 and 1.35 ha/ha
respectively), than Scenario (ii) and Scenario (i), which was limited by
the already high stocking rate of the “without project” situation, with
above average pasture management of 1.0 AU/ha in winter. It is notable
that, should fertilizer costs come down enough to make their use attractive
(not the present case), in scenario (iii) the incremental carrying capacity of
the land could reach 252 percent. Scenario (iv) shows an intermediate
position of the more likely situation for adoption with equal areas of crops

Vol. 5–2007 67
Fao_61_74.qxp 24-04-2004 13:09 Pagina 68


and pasture. This has considerable significance for the environmental
aspects of livestock development, firstly in the potential to mitigate
deforestation, but also in reduction of erosion losses from degraded
pastures through improvement in soil cover.


Total exploitable area (ha) 885.00 885.00 885.00 885.00
Degraded pasture area * 285.00 885.00 885.00 885.00
Stocking rate w/out project
AU/ha 1.00 0.50 0.50 0.50
AU/farm w/out project 285.00 442.50 442.50 442.50
Renovated pasture area 285.00 285.00 885.00 442.50
Renov. stocking rate AU/ha 1.76 1.76 1.76 1.76
AU/farm with project 501.60 501.60 1 557.60 778.80
Increase in AU/farm (fraction) 0.76 0.13 2.52 0.76
Hectares cropped 600.00 600.00 0.00 442.50
Increased crop yield 0.11 0.11 0.00 0.09
Crop area expansion 0.00 0.68 0.00 0.50
Mitigation potential (ha/ha)** 0.87 0.92 2.52 1.35
* Based on Case Study No. 1.
** Absorption of crop area is less efficient in mitigation than better pasture management including fertilization.
Note; if highly producing maize/soybeans replace low-yielding upland rice there could be a yield gain of 100%.
Source: Expanded from Landers and Weiss (2005).

Below are short descriptions of case studies 2–6 and a summary table
of their results.

Case study 2. A net area of 1 130 ha is farmed (not counting
environmental reserves). A rainfed maize and soybean rotation is
integrated with a cow/calf operation. ICLZT is introduced to rotate
pasture with crops, and the crop stubble in grazed during the dry season,
thus intensifying the cattle operation. After the summer crop, hay is
taken from volunteer grass and millet, to supplement the permanent
summer pasture, which becomes limiting in view of the ratio of 6.5:1 of
stubble grazing to permanent pasture. An expansion scenario is also

68 Integrated Crop Management
Fao_61_74.qxp 24-04-2004 13:09 Pagina 69


examined, using “without project” medium technology level and clearing
new land. Initial low soil fertility, light scrub savannah vegetation and flat
topography are typical of the Cerrado biome in this region bordering the
drier Northeast “sertão”. Case study 2 is situated in Formoso do Rio
Preto, West Bahia. Source: adapted from Landers and Weiss, 2005.

Case study 3. Weaned calves are brought in, reared on renovated and
subdivided ZT pasture and sold as steers (for fattening to slaughter
weight). A soybeans-soybeans-soybeans/maize arable phase follows the
pasture, with 3 ICLZT rotations (2 rainfed and 1 rainfed + irrigated) on
3 200 ha, including environmental reserves. The “without project”
situation uses a medium/high technology level without rotating crops
and pasture, the latter with large 200 ha paddocks. Initial soil fertility,
low, open scrub savannah vegetation, gently undulating topography,
Cerrado biome. Location: Cristalina, GO.

Case study 4. Integrated breeding/rearing/fattening pasture operation on
1 000 cleared ha; ICLZT is introduced to renovate old pastures with
mechanized crops, also providing dry season stubble grazing. The
“without project” scenario represents the clearing cycle in traditional
practice. Initial soil fertility moderate, originally heavy rainforest
vegetation, Amazon biome. Marabá, PA.

Case study 5. Small intensive beef cow/calf operation on 261 ha rotates
with ZT maize for silage and ZT pigeon pea in pastures. The latter and
ZT adoption are the sole changes in the operation. Old cleared area
(1940s), originally high soil fertility (old coffee land), Atlantic Forest
biome, Poloni, SP.

Case study 6. Medium-sized mechanized operation on 800 ha (excluding
reserves), with fattening of steers or heifers on a dark red clayey latosol
(Ferralsol), originally forest vegetation, at Maracajú MS at approximately
20º S latitude. Cropping pattern is a 1 year pasture ley in rotation with 3
years of annual cropping: soybeans-cotton-maize. Stocking rate on this
P. maximum pasture reaches 5 AU/ha, but if continued in the second year
this would drop to 3 AU/ha. Winter grazing consists of finger or pearl millet

Vol. 5–2007 69
Fao_61_74.qxp 24-04-2004 13:09 Pagina 70


and black oats, after soybeans; cotton is harvested late and is followed by
winter fallow. The finger millet is also used as a nurse crop for the Panicum,
to give early bite. No supplemental feed is used, except proteinized salt.
This system produces 3 hectares of arable crops and carries 5 AU per
4 hectares. To achieve the same production without integration of crops
and livestock, 3 hectares would be needed for annual crops plus another
10 hectares to maintain 0.5 AU per hectare in the customary grazing
system on degraded pasture cleared from forest. Four hectares of this
intensive integrated crop–livestock system thus save 9 hectares of forest
from clearance and land degradation. In short, 2.25 hectares of forest is
saved per hectare used.

From the case studies, it can be seen that the pasture phase (ley) after crops
varies from 1 to 6 years, depending on the intensity of the system. A one-
year ley, as in Case study 6, maximizes the carrying capacity of the pasture
and also the proportion of crops in the rotation. Crop phases with ICLZT
are between 3 and 4 years, allowing sufficient time for a significant build-
up in fertility, which is exploited by the pasture phase, at zero cost. In
return, the soil conditioning and breaking of pest, disease and weed cycles
during the pasture phase results in reduced costs and higher yields for the
crops. Under the traditional system of pasture establishment and
renovation, undersowing in an upland rice crop with about 250 kg/ha of
NPK fertilizer 4–14–8 for one year (maximum 2), generated far too little
residual fertility, which is the reason why most Brazilian tropical pastures
have fallen rapidly into degradation. The economic significance of the
cropping phase in an ICLZT system is that it permits cheap establishment
of high quality pasture and provides the mechanism to renovate it on a
sustainable basis, besides improving annual crop performance.
The simplest ICLZT system is that of Case study 2, using winter
grazing and haying of stubble to improve dry season carrying capacity.
This can be improved, as in Case study 1, by oversowing pasture seed
(usually in soya) or undersowing in maize or another nurse crop – these
pastures can either be annual or perennial grasses (even the annual
weed grass B. plantaginea could have application here). The former
and B. ruziziensis are preferred for oversowing, because they come away

70 Integrated Crop Management
Fao_61_74.qxp 24-04-2004 13:09 Pagina 71


faster. Case study 3 shows the greatest variety of rotations and uses
Brachiaria spp. to suppress kidney bean diseases in the non-irrigated and
irrigated rotations, and B. ruziziensis from seed as an annual catch crop
for grazing in the irrigated area, in the window between maize harvest
and bean planting. As shown in Case study 4, ICLZT has relevance for
the Amazon biome, but will only have significant uptake when clearing
new land becomes more expensive. Enriching permanent pastures with
pigeon pea, as in Case studies 1 and 5, or other legumes, is a good means
of improving their efficiency (alternatively, legume protein banks can be
used, but these have not had significant uptake in Brazil). The most
advanced system of grazing is that of the one-year ley in Case study 6,
which maximizes pasture stocking rates and the percentage of total farm
area in crops and hence, total farm income.
In Table 16 the results of economic analyses of the above case studies
are summarised (except Case study 6, included as an example of the most
intensive rotation possible).
The results of Table 16 indicate:
1. That the investment in intensifying to an ICLZT system is viable;
2. That investment levels are relatively modest;
3. That the highest impact of ICLZT is obtained with a high ratio of
winter grazing (on stubble or second crop areas) to permanent pastures;
4. That the highest stocking rates are achieved with a one-year ley.
All ICLZT systems show clear financial benefits in terms of IRR and
NPV, with the exception of Case study 4, where the inclusion of land sale
value in the expansion in the “without project” scenario inflates the IRR
when compared with the “with project” ICLZT scenarios, in which
renting land to a crop farmer shows a lower return than planting with
own machinery. However, these scenarios have attractive net returns and
very adequate IRRs as well.
Also, there is a large variation in Deforestation Mitigation Potential
(DMP), due to different proportions of crop to pasture and variations in
incremental stocking rates, with two extremes: Case study 5 has the
lowest proportion of crops and a high DMP of 2.52 ha/ha, but Case
study 6 has 75 percent of annual crops and a DMP of 2.25 (see Table 17),
due to the very high stocking rate allowed on a one-year ley. These
systems, especially the latter, will be the trend for the future.

Vol. 5–2007 71
Fao_61_74.qxp 24-04-2004 13:09 Pagina 72



(ha) (R$/ha) (%) R$/ha/yr AU/ha
Study No. 1, Cerrado – cow/calf,
annual rainfed/ irrigated crops
on 1 175 ha. total area. Med.
high mgt. – Goiás state. ICLZT 885 1.458 19.96 372 1.76

Without ICLZT. Without 885 1.433 14.16 229 1.05

Study No. 2, Cerrado – cow/calf,
annual rainfed crops, 1 130 ha.
Total area. ICLZT 820 1.000 8.85 84 2.18

400 ha expansion – crops not
rotated with pasture, cow/calf.
Med. mgt. – Bahia state. Expansion 1 220 1.340 5.38 83 0.52

Without expansion, without ICLZT. Without 820 933 3.30 53 0.52

Study No. 3, Cerrado – rainfed
crops x beef rearing high mgt. –
Goiás state. 3 200 ha total area. ICLZT Rot.1 2.400 2.598 19.63 187 1.49

Intensive grazing with electric
fences. ICLZT Rot.2 2.400 2.598 27.16 262 1.74

Irrigated crops. ICLZT Rot.3 2.400 4.098 26.71 301 1.93

Without rotation with pasture. Without 2.400 2.432 12.65 114 0.93

Study No. 4, Amazon – beef
Itinerant 10 yr clearing cycle
med. low mgt. – South of Pará Clearing
state. 1 000 ha of pasture. Scen.(i) 1 000 -559 35.7 37 0.98

Pasture renovation in situ med. Renov.
mgt. Scen.(ii) 1 000 805 9.72 203 1.22

ICLZT with rice/soybeans med. ICLZT
high mgt. Scen (iii) 1 000 698 11.68 310 1.60

Pasture renovation via renting ICLZT
20% of pasture – med. low mgt. Scen.(iv) 800 742 6.48 173 1.58

Study No. 5, Atlantic – small
farmer 261 ha. ZT + legumes in
pasture – med. high mgt – São
Paulo state. ICLZT 261 1.552 6.41 138 1.79

Conventional tillage, no 1.504 negative 16 1.20
legumes. Without 261

Study No. 6 With 800 n. a. n. a. n. a. 5.0
Without 800 n. a. n. a. n. a. 0.5
Source : Expanded from Landers et al. (2005).

72 Integrated Crop Management
Fao_61_74.qxp 24-04-2004 13:09 Pagina 73


An analysis of DMP in Table 17 compares the effects of the best
alternative management systems on this parameter (taking the best
scenarios of those analysed). The land use intensification through
increased stocking and absorption of crop expansion in areas of degraded
pasture allows mitigation of land clearing on the scale of approximately
0.25 to 2.5 hectares for every hectare in ICLZT, depending on the system.
It can be seen that the two most important variables are the increase in
stocking rate in the pasture and the proportion of pasture to crops.


PARAMETER No. 1 No. 2 No. 3 No. 4 No. 5 No. 6
Degraded pasture area* (ha) 885.00 980.00 2 400.00 1 000.00 252.00 800.00
AU/farm w/out project 442.50 490.00 1 200.00 500.00 126.00 400.00
Renovated pasture area (ha) 285.00 150.00 1 200.00 800.00 241.00 200.00
Renov. stocking rate AU/ha 1.76 1.96 2.03 1.22 1.82 5.00
AU/farm with project 501.60 294.00 2 436.00 780.80 439.00 1 000.00
Increase in AU/hectare 0.13 -0.40 1.03 0.56 2.48 1.5
Crop area (proportion) 0.68 0.85 0.50 0.20 0.04 0.75
Mitigation potential (ha/ha)** 0.81 0.45 1.53 0.76 2.52 2.25
* Assumes that the farm starts as 100% degraded pastures, with a stocking rate (winter) of 0.5 AU/ha.
** Crop yield increases (decreases) and reduced time to sale of animals were not considered and are much smaller
effects. These factors are almost invariably positive, thus these assumptions are conservative.

Source : Landers et al. (2005).

Vol. 5–2007 73
Fao_61_74.qxp 24-04-2004 13:09 Pagina 74
Fao_75_94.qxp 7-05-2004 15:32 Pagina 75

Sustainable agriculture
and policy considerations

A sustainable approach to agriculture calls for a system that is not only
environmentally friendly, but economically viable and socially responsible
as well. Important allies for such an approach therefore include the CA
farmers, but also the governments and societies that support these farmers.
The environmental awareness of a CA farmer is considerably enhanced and
his/her operation more profitable – mostly through ZT, and especially by
increasing land use intensity (LUI) through ICLZT. However, societal
investments in the transition to CA are important catalysts in the
sustainable agriculture movement. It is important to recall that farmers
worldwide have considerably reduced the cost of basic foods over the last
decades (Vieira et al., 2001), and with the increased efficiency of ZT, a net
transfer of value is generated to the consumer. In addition, food security,
biodiversity conservation and the sustainable use of natural resources are
common concerns for each member of society, so conversion investments
to CA, ZT and ICLZT will be important agenda items for many national
governments. Many of these governments need assistance with the
promotion of conversion to such sustainable systems, which will result in
economic gains for small farmers while increasing important environmental
services for society as a whole. Therefore, funding by international donors
and donor countries for programmes to extend CA, ZT and ICLZT in the
tropics and subtropics will need to be continued and expanded, with
technical support by international and national organizations such as
A CIRAD project in the mountainous areas of Vietnam has promoted the inclusion of forage production in
cropping systems as a means for adopting direct seeding systems that successfully integrate livestock. This
project has contributed to increased food and feed system performance, soil improvement, and forest
conservation (Husson et al., 2006).
Direct seeding, mulch based cropping systems in the humid tropical east coast of Malagasy, supported by
ANAE have contributed to soil conservation, fodder production, reduced forest clearing and the intensified
sustainable production of annual crops, leading to improved socio-economic conditions for small farmers
(Séguy et al., 2006).
Besides medium and large farms, thousands of smallholders in Paraguay are practising conservation
agriculture, with support from GTZ, which has resulted in improved incomes and labour savings; these tend
to be reinvested in intensification and diversification of their production systems (Lange, 2005).

Vol. 5–2007 75
Fao_75_94.qxp 7-05-2004 15:32 Pagina 76


Conservation agriculture shows a clear trend in increased yields and
reduced use of agricultural chemicals; and even if their amount would
remain constant, their use per tonne of product would be reduced,
thereby saving farmers money. In addition, CA farmers tend to have
more time for their families and for professional training, and larger CA
farmers generally train their personnel and pay higher wages.
The case studies on ICLZT systems in the Brazilian Amazon,
Savannah and Atlantic Forest biomes described in chapter 5 were selected
in order to determine what kinds of incentives might be necessary for the
farmer to change his/her economic decision from land clearing to land
intensification. Results showed that ICLZT technology would allow
productive annual crops to be grown on a significant proportion of the
approximately 80 million hectares of cultivated pastures present in the
Amazon and Cerrado biomes today, of which 70–80 percent are classified
as degraded (Sano et al., 1999; Vilela et al., 2004).
In twenty-year investment models, the study shows increased long-
term financial returns to ICLZT (improved NPV, net return/ha and IRR),
but incremental investments in the transition and new management skills
are needed to achieve the best returns. The expression, “a farmer in the red
cannot look after the green” is especially poignant here, since many small
farmers are not able to make the needed one-time investments without
some support. When farmers improve their levels of land stewardship,
they merit environmental services payments, not just because of the extra
costs involved in the transition, but also as a quid pro quo for the reduced
off-farm impacts and consequent public savings generated.

Zero tillage substitutes for, or reduces the use of, many leachable herbicides
used in conventional tillage, such as Trifluralin and Atrazine (a very soluble
maize herbicide used widely in conventional tillage), that have recognized
pollutant effects. Other chemicals used are the same as for conventional
tillage, but the environmental conscience awakened in CA farmers
increases the adoption of integrated pest management (IPM), ensuring
reduced use and safer handling and application of these products. Most

76 Integrated Crop Management
Fao_75_94.qxp 7-05-2004 15:32 Pagina 77


modern selective post-emergent herbicides are applied at very low dosages
of active ingredients, (often only grams/ha) and these, along with all other
agricultural chemicals, are subject to intense biological breakdown by the
action of the bacteria and fungi which feed on the surface humus, besides
being bound to soil organic matter and mineral surfaces. Zero tillage
systems avoid Atrazine or reduce its use to post-emergence, thereby
creating a ZT buffer effect, whereas pre-emergence herbicide use is the
norm in most conventional systems.
Desiccant herbicides, whose use in ZT have given rise to some
criticism, are either practically unleachable because they are bound to soil
colloids within minutes of application (Prata et al., 2000; Goellner 1989),
which is the case of Glyphosate, or have a very short half-life due to
biodegradability or instability, as for Ammonium Gluphosinate. Table 18
shows that the principal desiccants are alone in the least leachable
category of the groundwater ubiquity score (GUS). For example,
Glyphosate, by far the most widely used desiccant in ZT systems, is
classified in the lowest toxicology rating in Brazil (Class IV, green stripe),


(ml/g) (time in days) GROUNDWATER
Glyphosate 24.000 47 -0,6357 Very low
Ammonium Gluphosinate 100 7 1.6902 Low
Paraquat 1 000 000 1.000 -6.0000 Very low
Alachlor 124 21 2.5209 Medium
Atrazine 100 60 3.5563 High
Chlorimuron 110 40 3.1378 High
Cyanazine 190 14 1.9727 Medium
2,4-D DMA salt 20 10 2.6990 Medium
Diuron 480 90 2.5773 Medium
Fomesafen 60 100 4.4437 Very high
Metholachlor 200 118 3.5201 High
Nicosulfuron 30 21 3.3358 High
* Organic carbon partitioning coefficient (used for measuring how tightly bound a pesticide is).
** Groundwater ubiquity score: Calculated by the formula GUS = log T 1/2 x (4 – log Koc); values below 1.8
indicate low potential for groundwater contamination.

Source: Adapted from the Herbicide Handbook, SSA, 7th Edition (1994).

Vol. 5–2007 77
Fao_75_94.qxp 7-05-2004 15:32 Pagina 78


based on its very low toxicity to humans and non-leachability. This,
however, does not mean that these products can be used carelessly.
Farmers applying CA mainly use crop rotations, cover crops and
mechanical means like the knife roller for weed management, and use
desiccants and herbicides only if and as needed.
On CA farms, proof of enhanced biological activity abounds,
principally the increased levels of soil fauna and flora, of which the
mesofauna are the most evident, exemplified by earthworms and the
burrows of scarab beetle larvae. Hayes et al. (2003) showed that
concentrations of as little as 0.1 ppb of Atrazine have been shown to
cause hermaphroditism and genital deformity in frogs on conventional
farms, and although Atrazine-resistant populations are now appearing,
amphibian population decline continues to be a serious global matter.
According to the World Conservation Union (IUCN, 2001),
agricultural lands continue to be an important focus for biodiversity
conservation, and one of the keys to continued conservation is
ecological education of decision-makers and farmers, which allows them
to continue adopting more environmentally friendly farming methods.
By limiting the amount of chemicals released into air, soil and water, CA
farmers can continue to play a crucial role in biodiversity conservation
by the protection of specific habitats.
In addition to the reduction of herbicides used, the increased
environmental awareness of CA farmers in Brazil has led to a number of
farmers’ groups (Clubes Amigos da Terra) managing, or actively
collaborating with, centres for the recycling of used agricultural chemical
containers. In 2005 alone, Brazil recycled approximately 65 percent of
these containers (INPEV, 2005).
Compared to conventional agriculture using ploughs and discs, ZT
actually improves soil quality over time, reversing the loss of SOM. In the
pampas of Argentina, which were originally fertile soils, top yields were
produced without fertilizer, but SOM has declined from 6 percent a
century ago to 2 percent today, as a result of exploitation with
conventional land preparation, provoking continuous oxidation of
organic matter, in order to mine soil fertility and generate agricultural
exports. The pampas example further underlines the need for a sustainable
approach to modern agriculture, such as through ZT and ICLZT.

78 Integrated Crop Management
Fao_75_94.qxp 7-05-2004 15:32 Pagina 79


The fundamental changes in soil behaviour and management, referred to
in Chapter 1, bring a number of benefits to society, besides reducing on-
farm production costs (Landers et al., 2001). Adjusting the approximate
first estimate for Brazil (which requires revision, but used the most
conservative data available at the time) for 50 percent versus 35 percent
of adoption, the total gains with CA were estimated at US$ 1.9 billion in
2004/5. Besides the direct benefits to farmers (26 percent of the total),
the principal indirect-use benefits constitute reductions in public
spending. These derive from the value of the reduced clearing of native
vegetation (57 percent of the total); reduced off-farm effects of soil
erosion (silting, road maintenance and water treatment); lower
emissions of greenhouse gases; carbon sequestration; and enhanced
aquifer recharge (due to increased rainfall infiltration).
Besides these quantifiable benefits, a number of un-quantified
benefits have been noted after the conversion to CA, usually increasing
over time. Dust clouds from land preparation are eliminated in rural
towns, as are the effects of residue burning on air quality; flood peaks
are lower; and base flows of watercourses increase and springs have
begun to flow again due to the 70 percent greater rainfall infiltration (De
Maria, 1999). Also, the cost of reservoir silting may have been seriously
underestimated, since another estimate (Carvalho et al., 2000) was 76 times
higher than the conservative value chosen by Landers et al. (2001).
These assumptions require more in-depth study, which may well
increase the recognized benefits of converting to CA, ZT and ICLZT
systems. It is implicit that not paying environmental services for
preservation of native vegetation is a negative stimulus to conservation.
In many tropical and subtropical regions, land clearing is a response
to increased demand for agricultural products. Land clearing, in turn,
leads to loss of biodiversity, which is a growing concern in many
societies. But without a value placed on the preservation of natural
vegetation, land clearing will continue to be the cheapest route to
increase food and fibre production. However, if environmental services
payments were offered for increased LUI, deforestation could become
economically less attractive.

Vol. 5–2007 79
Fao_75_94.qxp 7-05-2004 15:32 Pagina 80


Table 17 in Chapter 5 shows, in the bottom line, how many hectares of
decreased demand for clearing native vegetation correspond to each
hectare of LUI with ICLZT, starting from a base of degraded pasture. The
realization of this mitigation potential depends on effective policy stimuli
to ICLZT and constraints to clearing; the former implies a value put on
each hectare of native vegetation spared, to use as a base for an
environmental services payment to the farmer who intensifies the use of his
land, thereby eliminating the need for frontier expansion. The latter
implies tighter control of licensing to clear new land and effective policing
of this, which will be enormously facilitated if farmers feel they are being
fairly treated with incentives to LUI. With the right incentives it is possible
to address this problem, as studies show that farmers respond logically to
economic stimuli, thereby making LUI more attractive than clearing.
The mitigation potential of each system in Table 17 ranges from 2.52 ha
of reduced demand for clearing new land for every hectare in the ICLZT
system of Model No. 5, to 0.45 ha/ha in Model No. 2. The average of all
models and their variants adopting ICLZT was 0.99 ha saved from
clearing per ha converted to an intensified integrated crop–livestock ZT
system. The differences are due to different proportions of annual crops
to pastures (stubble grazing to permanent pasture) and different increases
of stocking rates over the “without-project” situation. This shows
considerable potential for policy actions to mitigate the loss of native
vegetation, if society were willing to contribute to the transition costs of
LUI and supply technical capacity-building and credit for the
investments required. The largest of these is a relatively short-term
increase in incremental working capital for purchase of additional
animals to take advantage of the greater forage availability.

As a result of minimizing erosion, increasing, or even maximizing, crop
yields, soil biological activity and nutrient recycling, reducing market and
climate risks and improving profits, farmers in Brazil are focusing on
ICLZT, which intensifies land use and, consequently, mitigates the demand
for further land clearing. This win-win-win situation merits the recognition
of society, through a policy of both financial and non-financial incentives.

80 Integrated Crop Management
Fao_75_94.qxp 7-05-2004 15:32 Pagina 81


Farmers in other tropical and subtropical regions have also started
focusing on ICLZT, but reliable support facilities to aid in the transition
appear to be the key to conversion from conventional farming systems to
intensive ZT systems, especially for small farmers. Without the
appropriate incentives, training, or equipment, many small farmers may
see conversion as too risky an undertaking.

Addressing the conversion needs of small farmers
In addition to training and up-to-date information on ICLZT systems,
more tangible items such as cheap and simple equipment are considered the
first step in addressing the needs of small farmers wishing to convert.
Examples of simple and easily adaptable equipment, which can be used
manually or drawn by animals, can be seen in Plates 19–27 on the following
pages. Institutions or extension services can support small farmers by
providing equipment or micro-loans needed to purchase the equipment, as
well as the necessary training and information needed to begin.
Plates 19–27 were chosen from a photo gallery of FAO-supported
projects to illustrate equipment that can be adapted for resource-poor
farmers. Their use in different regions shows that CA, ZT and ICLZT
systems are adaptable for a range of biophysical environments.

PLATE 19: Hand jab planter demonstrated for direct seeding by hand. Guantanamo, Cuba. October, 2006.

Vol. 5–2007 81
Fao_75_94.qxp 7-05-2004 15:32 Pagina 82


PLATE 20: Manual traction boom sprayer used PLATE 21: Animal traction no-till planter used for
for desiccation of degraded Brachiaria pasture direct planting of maize into knife-rolled and
as preparation for no-till maize planting. desiccated spontaneous vegetation (weeds) used
Goiania, Brazil. September, 2006. as soil cover. Posoltega, Nicaragua. June, 2005.

PLATE 22: Animal traction boom sprayer used for desiccation of cover crop. Posoltega, Nicaragua.
June, 2005.

PLATE 23: Animal traction no-till planting of PLATE 24: Animal traction rolling punch
maize on degraded hill site. Jung San Up planter made in Kenya for direct seeding into
farm, Democratic Peoples Republic of Korea. unprepared soil. Jinja, Uganda. May, 2002.
June, 2004.

82 Integrated Crop Management
Fao_75_94.qxp 7-05-2004 15:32 Pagina 83


PLATE 25: Animal traction no-till planter used for planting beans into desiccated sorghum.
Guantanamo, Cuba. October, 2006, (details in Plates 26 and 27).

PLATE 26: Chisel-type fertilizer boot (furrow PLATE 27: Close-up view of the cutting disc,
opener) behind cutting disc; offset double-disc fertilizer boot and offset double-disc seed
seed furrow opener at rear. Brazilian type animal furrow opener (of a tractor-mounted ZT
traction no-till planter, used in various countries. planter).

Equipment suitable for small farmers as illustrated in Plates 19–27,
which should be easily adaptable for a wide range of biophysical
conditions, is necessary for promoting ICLZT systems in the different
tropical and subtropical regions of the world. The most important
concept to keep in mind is that while the principles of CA, ZT and
ICLZT are universal, the solutions are local and farmer-led (which crops
to rotate, which equipment to use, access to inputs, etc.). Therefore,

Vol. 5–2007 83
Fao_75_94.qxp 7-05-2004 15:32 Pagina 84


institutions or extension services supporting small farmers must inquire
from the farmers themselves about the bottlenecks they are facing. They
should assure availability and accessibility of low-cost equipment,
facilitated and supported by local and national government actions and
enabling policies. Technical solutions to local problems during the
adoption phase have to be found in collaboration among farmers,
extension technicians and scientists. The classic roles of technology
development and transfer have proven not to work well for CA and to
take too much time. The farmer him/herself is often the most creative
developer when it comes to finding locally adapted solutions. Technical
support from organizations such as FAO and CIMMYT, EMBRAPA
and CIRAD, and funding from the World Bank, Regional Development
Banks and donor countries are especially important to further encourage
governments to facilitate the spread of ICLZT.

In the drive to meet the demands of present-day society, without
compromising the ability of future generations to meet their own needs,
sustainable forms of agriculture are an essential element. Farmers are all
too often caught in the cross-fire of differing societal demands for
products that are cheap, yet produced in an environmentally friendly
and socially responsible way. Since CA, ZT and ICLZT systems as
described in this publication address some important societal needs such
as food security, biodiversity preservation and natural resource
conservation, they constitute a major step towards systems that can
truly be called sustainable.
While the specific technologies vary depending on local conditions, the
approaches toward sustainable intensification based on incorporation of
pastures and livestock into ZT crop production are very promising. The
biology of the approaches of what is found to work in Brazil will give
useful insights on what farmers in other parts of the world, including
smallholders in African savannahs, might consider when developing their
own sustainable intensified production systems. This publication can
therefore be used as a starting reference, benefiting not only farmers and
consumers in tropical and subtropical regions, but possibly expanding into
other regions as well, thus serving to benefit human society as a whole.

84 Integrated Crop Management
Fao_75_94.qxp 7-05-2004 15:32 Pagina 85


Policy recommendations for promoting ICLZT in tropical and
subtropical regions
• Equipment lending, especially to small farmers
• Providing farmers with seed for cover crops, including through
producer training and market development
• Support to the input supply sector to make suitable ZT equipment
and inputs available and accessible to farmers in the long term
• Promotion and funding of farmers clubs
• Micro-loans enabling small farmers to convert to CA and diversify
• Reduced premiums on crop insurance
• Payments for environmental services performed (such as erosion
control, carbon sequestration, forest/biodiversity preservation,
watershed protection and habitat conservation)
• Increased cooperation among governments, agricultural extensionists
and research institutes
• Increased information exchange networks between national and
international research organizations, development organizations,
private companies, producers and cooperatives, and donors
• Further funding of ICLZT projects by international donors and
donor countries.

Vol. 5–2007 85
Fao_75_94.qxp 7-05-2004 15:32 Pagina 86
Fao_75_94.qxp 7-05-2004 15:32 Pagina 87


Amado, T.C. & Costa, C.N. 2004. Solos sob sistema Plantio Direto no Brasil
podem atuar como importante tampão ambiental. Jornal Direto no Cerrado
No. 37: 21–22. Associação de Plantio Direto no Cerrado, Brasília, D.F.
Baker, C.J. 2006. Seeding openers and slot shape. In C.J. Baker & K.E. Saxton,
eds.No-tillage seeding in conservation agriculture, second ed., pp. 34–59.
Rome, FAO; and Wallingford, UK and Cambridge, MA, CABI.
Bernardi, A.C.C.; Machado, P.L.O. de A.; Freitas, L. de; Coelho, M. R.;
Leandro, W.M.; Júnior, J.P.O.; Oliveira, R. P.; Santos, H.G. dos; Madari, B.E.
& Carvalho, M.C.S. 2003. Correção do Solo e Adubação no Sistema de
Plantio Direto nos Cerrados. Rio de Janeiro: EMBRAPA solos,
Documentos, No. 46. 22 pp.
Blancaneaux, P.H.; Freitas, P. L. de; Amabile, R. F. & Carvalho, A.M. de. 1993.
Le semis direct comme practice de conservation des sols du Brésil Central.
Cahiers ORSTOM, série Pédologie, 28: 253–275.
Boddey, R.M.; Macedo, R.; Tarré, R.M.; Ferreira, E.; Oliveira, O.C. de;
Rezende, C. De.; Cantarutti, R.B.; Pereira, J.M.; Alves, B.J.R. & Urquiaga,
S. 2004. Nitrogen cycling in Brachiaria pastures: The key to understanding
the process of pasture decline? Agriculture, Ecosystems and Environment,
103: 389–403.
Bonamigo, L.A. 1999. Recuperação das Pastagens com Guandu em Sistema de
Plantio Direto. In: Plantio Direto na Integração Lavoura Pecuária, Cabezas,
W.A.R.L.; Freitas, P.L. Instituto de Ciências Agrárias da Universidade
Federal de Uberlândia e Associação de Plantio Direto no Cerrado.
Broch, D.L.; Pitol, C. & Borges, E.P. 1997. Integração Agricultura – Pecuária:
Plantio Direto da soja sobre pastagem na integração agropecuária. Maracaju –
MS, FUNDAÇÃO MS para pesquisa e Difusão de Tecnologias
Agropecuárias. 24 pp. (Informativo técnico, 01/97).
CAT Uberlândia. 2005. Website (
Camargo, G.S.D.; Zen, S.D.; Ishihara, A.M.; Osaaki, M. & Ponchio, L.A.
2002. Economia da pecuária de corte e o processo de ocupação da
Amazônia. Manuscript. Centro de estudos Avançados em Economia
Aplicada–CEPEA–ESALQ/USP, Piracicaba – SP, Brazil.
Carvalho, N.O., Guilhon, L.G. & Trindade, P.A. 2000. Sedimentation of a
small reservoir – Itiquira, a case study (in Portuguese). Rev. Bras. Rec.
Hídr., 5(1): 69–79.
Cobucci, T. & Portela, C.M. de O. 2003. Manejo de Herbicidas no Sistema
Santa Fé e na Braquiária como Fonte de Cobertura Morta. In Kluthcouski
et al., eds. 2003.

Vol. 5–2006 87
Fao_75_94.qxp 7-05-2004 15:32 Pagina 88


Constantin, J. & De Oliveira Jr, R.S. 2005. Dessecação antecedendo a semeadura
direta pode afetar a produtividade. Informações Agronômicas No. 109:
14–15. Potafós, Piracicaba, SP
Da Costa, J.L. & Rava, C.A. 2003. Influência da Braquiária no Manejo de
Doenças do Feijoeiro com Origem no Solo. In Kluthcouski et al., eds. 2003.
De Maria, J.C. 1999. Erosão e Terraços em Plantio Direto. Boletim Informativo
da Sociedade Brasileira.
Dos Santos, E.V. 2003. A Nova Contribuição do Sistema de Plantio Direto.
Revista Plantio Direto Nº 76: 28–34.
Döwich, I. 2002. Estratégias e desafios à adoção permanente do sistema plantio
direto. In VI encontro Plantio Direto.
FAO. 2001. Zero tillage development in tropical Brazil. The story of a successful
NGO activity, by J.N. Landers. FAO Agricultural Services Bulletin No.
147. Rome.
FAO. 2006. Conservation agriculture homepage. (
Gianluppi, D. 1999. O uso de PD nos cerrados de Roraima. Jornal Direto no
Cerrado. Year 4, No. 11: 9. Associação de Plantio Direto no Cerrado,
Goiânia, Goiás, Brazil.
Goellner, C.I. 1989. Ecotoxicologia e toxicologia do herbicida paraquat. Gráfica e
Editora UPF, Passo Fundo-RS, Brazil. 93 p. ilust.
Goedert, W.J. 1985. Solos dos Cerrados – Tecnologias e estratégias de manejo.
Editora Nobel. Brasília: EMBRAPA, Centro de Pesquisa Agropecuária dos
Cerrados, Planaltina, DF, Brazil.
Hayes, T.; Haston, K.; Tsui, M.; Hoang, A.; Haeffele, C. & Vonk, A. 2003.
Atrazine-induced hermaphroditism at 0.1 ppb in American Leopard Frogs
(Rana pipiens): Laboratory and field evidence. Environmetal Health
perspectives, 111(4).
Homma, A.K.O.; Carvalho, R.A.; Sampaio, S.M.N.; Da Silva, B.N.R.; Silva,
L.G.T. & De Oliveira, M.C.C. 2001. A Instabilidade dos Projetos de
Assentamentos como Indutora de Desmatamentos no Sudeste Paraense. 4º
Reunião Nacional Ecologia x Economia, Belém, PA, 2001. CD ROM Ass.
Bras. De Ecolog. Econ., Campinas, SP.
Husson, O.; Chabanne, A.; Ha Dinh Tuan; Lecomte, P.; Martin, C.; Castella,
J.C.; Tivet, F. & Séguy, L. 2006. Integrating crops and livestock through
direct seeding on vegetal cover in Vietnam. In 2001-2005 Cirad’s contribution
to the World Congress on Conservation Agriculture (available at
IBGE. 1996. Censo Agropecuário. Instituto Brasileiro de Geografia e Estatística,
Rio de Janeiro, Brazil. (
INPE 2006. Website. (
INPEV 2005. Website. (
IUCN. 2001. Influencing agriculture policy for biodiversity conservation. IUCN
European Programme Report. 263 pp. (also available at

88 Integrated Crop Management
Fao_75_94.qxp 7-05-2004 15:32 Pagina 89


IUSS (International Union of Soil Sciences) Working Group WRB. 2006. World
Reference Base for Soil Resources 2006. World Soil Resources Report No.
103. FAO, Rome.
Kaimowitz, D. & Angelsen, A. 2001. Agricultural Technologies and tropical
deforestation. Wallingford, UK, CABI.
Kluthcouski, J.; Oliveira, I.P.; Yokoyama, L.P.; Dutra, L.G.; Portes, T. De A.;
Silva, A.E. Da; Pinheiro, B.S.; Ferreira, E.; Castro, E. Da M. De &
Guimaraes, C.M. 1999. Sistema Barreirão: recuperación/renovación de
pasturas degradadas ultizando cultivos anuales. In E.P. Guimarães, J.I.
Sanz, I.M. Rao, M.C. Amézquita & E. Amézquita, eds. Sistemas
agropastoriles en sabanas tropicales de América Latina, pp. 195–231. Cali,
Colombia, CIAT & EMBRAPA.
Kluthcouski, J., Stone, L.F. & Aidar, H., eds. 2003. Integração lavoura-pecuária.
Santo Antonio de Goiás, Brazil, EMBRAPA Arroz e Feijão. 570 pp.
Landers, J.N., ed. 1994. Fasciculo de Experiencias de Plantio Direto no Cerrado.
Goiânia, Brazil, APDC. 254 pp.
Landers, J.N. 1999. Technology transfer mechanisms for the new zero tillage
techniques in the savannahs of central Brazil and its benefits for the
environment. In F. Rasolo & M. Raunet, eds. Gestion Agrobiologique du
Sol et des Systèmes de Culture - Acts of the international workshop,
Antsirabe Madagascar 1998. Montpellier, France, CIRAD.
Landers, J.N.; Sant’ana, G.; Rocha, M.T.; Manfrinato, W. & Weiss, J. 2001.
Environmental impacts of zero tillage in Brazil – a first approximation. In L.
García-Torres, J. Benítez & A. Martínez-Villela, eds. Conservation agriculture:
a worldwide challenge, pp. 317–326. Madrid, ECAF and Rome, FAO.
Landers, J.N. & Weiss, J. 2005. Study on the conversion of degraded tropical
pastures to productive crop x livestock rotations and their effect on
mitigating deforestation. Consulting report to WWF-USA and the Nature
Conservancy. 30 pp. (in press).
Landers, J.N.; Weiss, J. & Clay, J. 2005. A new solution for old problems in
Brazil: crop–livestock rotations with zero tillage as a sustainable land
management tool. Proceeding of: The International Symposium on Land
Degradation and Desertification, Uberlândia, MG, Brazil, May 2005.
Universidade Federal de Uberlândia, Uberlândia, Brazil. (in press).
Lange, D. 2005. Economics and evolution of smallholdings conservation
agriculture in Paraguay. Mid-term experiences. 92 pp., appendix on CD.
FAO; GTZ; Prog. Nac. de Man., Cons. y Rec. de Suelos, MAG (Min. Agric.
and Livestock Mgt), Asunción, Paraguay.
Lara Cabezas, W.R. 1998. Nitrogênio em cobertura – caminhos e áreas cinzas.
Jornal Direto no Cerrado, Ano 2, No. 5, Jan/Mar. 1997, Associação de
Plantio Direto no Cerrado, Giânia, Goiás, Brazil.
Lara Cabezas, W.R. 2003. Sobressemeadura com sementes de milheto revestidas
no Triângulo Mineiro-MG: estudo preliminar. Núcleo de Solos,
Universidade Federal de Uberlãndia, Uberlândia, Minas Gerais, Brazil.
Unpublished data.

Vol. 5–2007 89
Fao_75_94.qxp 7-05-2004 15:32 Pagina 90


Lara Cabezas, W.R. & Freitas, P.L. de, eds. 1999. Plantio Direto na Integração
Lavoura Pecuária. Universidade Federal de Uberlândia. Uberlândia, MG,
Lopes, A.S. 2004. Sistema Plantio Direto: bases para o manejo da fertilidade do
solo. São Paulo, ANDA.
Macedo, M.C.M. 1997. Sustainability of pasture production in the savanas of
tropical America. International Grassland Conference, 18. Proceedings 4:
391–399. Canadian Forage Council, Canadian Society of Agronomy,
Canadian Society of Animal Science, Winnipeg and Saskatoon.
Macedo, M.C.M. 1999. Degradação de pastagens: conceitos e métodos de
recuperação. Simpósio sobre a sustentabilidade do leite no Brasil, Goiânia-
GO Brazil. Anais: Embrapa Gado de Leite, Juiz de Fora – MG, Brazil pp.
Machado, A.B.; Neto, M.B.R.; Pereira, P.G.P.; Caldas, E.F.; Gonçalves, D.; Santos,
N.S.; Tabor, K. & Steininger, M. 2004. Estimativas da Perda da área do
Cerrado brasileiro. pp. 23. Conservaçãio Internacional, Brasília DF, Brazil.
Magnabosco, C., de U.; Los Reyes, C.B.A.; De Faria, C.U. & Barcellos, A. de O.
2002. Avaliação da contribuição do componente genético no crescimento ao
ano e sobreano de bovinos da raça nelore, recriados em pastagens renovadas
no Cerrado. Folder, EMBRAPA Centro de Pesquisas de Arroz e Feijão,
Santo Antonio-GO, Brazil 11 pp.
Mahar, D.J. 1989. Government policies and deforestation in Brazil’s Amazon
region. Washington, DC, World Bank. 156 pp.
Martha Junior, G.B.; Barioni, L.G.; Vilela, L. & Barcellos, A.O. 2004.
Dimensionamento da Área de Pastagem e da Taxa de Lotação no Pastejo
Rotacionado. Comunicado Técnico, EMBRAPA Cerrados Planaltina (in press).
McClung, A.C.; Freitas, L.M.M.; de Gallo, J.; Romano, R.; Quinn, L.R. & Mott,
G.O. 1958. Prelimnary Fertility Studies on “Campos Cerrados” soils in Brazil
(English edition). New York, IRI Research Institute.
Mikkelsen, D.S., De Freitas, L.M.M. & McClung, A.C. 1963. Effects of Liming
and Fertilization on Cotton, maize and Soybeans on Campo Cerrado Soils –
State of Sao Paulo, Brazil. New York, IRI Research Institute.
Nachtergaele, F.O. & Brinkman, R. 1996. Identification of analogous land for
agrotechnology transfer in the savannah zones of the developing world.
Integrated Crop Management, 1: 29–42.
Nasser, L.C.B.; Spehar, C.R.; Saueressig, M.G. & Freitas, M.A. 2001. Controle de
Mofo Branco em Lavouras e Pastagens no Cerrado. Jornal Direto no
Cerrado. Ano 6. No. 22: 10.
Oliveira, J.H.F. de. 2002. Nelore: Base Genética e Evolução Seletiva no Brasil.
Planaltina, DF, EMBRAPA Cerrados. 54 pp.
Prata, F.; Lavorenti, A.; Regitano, J.B.; Tornisielo, V.L. 2000. Influência da
Matéria Orgânica na Sorção e Dessorção de Glicofosfato em Solos com
Diferentes Atributos Mineralógicos. Revista Brasileira de Ciência do Solo 24
(4), outubro a dezembro de 2000. Viçosa – MG, Brazil.

90 Integrated Crop Management
Fao_75_94.qxp 7-05-2004 15:32 Pagina 91


Quinn, L.R.; Mott, G.O.; Bisschoff, W.V.A. & Da Rocha, G.L. 1963. Beef
Production of Six Tropical Grasses. New York, IRI Research Institute.
Sá, J.C. de M. 1993. Manejo da fertilidade do solo no Plantio Direto. Fundação
ABC, 1993. Castro, Paraná, Brazil. 96 pp.
Sá, J.C. de M.; Cerri C.C.; Piccolo M.C.; Feigl, B.E.; Buckner, J.; Fornari, A.; Sá,
M.F.M.; Seguy, L.; Bouzinac S.; Venzke-Filho, S.P.; Paulleti, V. & Neto, M.S.
2004. O Plantio Direto como base do sistema de produção visando o
seqüestro de carbono. Revista Plantio Direto No. 84: 45–61.
Sano, E.E., Barcellos, A.O. & Bezerra, H.S. 1999. Área e distribuição espacial de
pastagens cultivadas no Cerrado brasileiro. Boletim de Pesquisa No. 3.
EMBRAPA Cerrados, Planaltina, DF.
Santana, D.P.; Pereira Filho, I.A.; Sans, L.M.A.; Cruz, J.C. & Alvarenga, R.C.
1994. Determinações de perdas de solo e água sob diferentes condições de
manejo, em um podzólico Vermelho-Amarelo de Sete Lagoas, MG. Relatório
Técnico Anual do Centro Nacional de Pesquisa de Milho e Sorgo. Período
1992-1993. Sete Lagoas, 1994. 6: 319–320.
Saturnino, H.M. & Landers, J.N., eds. 2002. The environment and zero tillage.
Associação de Plantio Direto no Cerrado, Brasília. 144 pp.
Scalea, M. 1999. A Cultura do Milheto e seu uso no Plantio Direto no Cerrado. In
F. Farias Neto Workshop Internacional do Milheto, Planaltina, DF: EMBRAPA.
Séguy, L. & Bouzinac, S. 1997. Pequeno Guia da gestão das coberturas para os
sistemas de cultura mecanizada em Plantio Direto. CIRAD, Rhodia S.A.,
Grupo Maeda and Cooperlucas, Goiânia, Goiás, Brasil.
Séguy, L.; Bouzinac, S.; Maronezzi, A.C.; Taffarel, W. & Taffarel, J. 1998. O
Plantio Direto de arroz de sequeiro de alta tecnologia na zona tropical úmida
do Centro-Norte do Mato Grosso, Brasil. CIRAD, Grupo Maeda, Agronorte,
Goiânia, Goiás, Brasil.
Séguy, L. & Bouzinac, S. 2002. Gestão dos solos tropicais sob Plantio Direto. pp
121–134 in: Anais do 7º Encontro de Plantio Direto no Cerrado. Clube
Amigos da Terra do Oeste Baiano, Luiz Eduardo Magalhães, Bahia, Brazil.
Séguy, L.; Bouzinac, S.; Scopel, E.; & Ribeiro, F. 2006. New concepts for
sustainable management of cultivated soils through direct seeding mulch based
cropping systems: the CIRAD experience, partnership and networks. In: 2001-
2005 Cirad’s contribution to the World Congress on Conservation Agriculture.
(available at
Smith, M.F. 2004. Exemplo de como profissionalizar a pecuária. Jornal Direto no
Cerrado Ano 9, No. 38: 10. Associação de Plantio Direto no Cerrado,
Brasília, D. F., Brazil.
Sousa, D.M.G. & Lobato, E. 2003. Adubação Fosfatada em Solos da Região do
Cerrado. Informacões Agronômicas. Nº 102, Junho 2003. POTAFOS,
Piracicaba, São Paulo, Brazil. pp 1-16.
Stone, L.F, & Moreira, J.A.A. 1998. A irrigação no plantio direto. Jornal Direto
no Cerrado, Ano 3, No. 8. Associação de Plantio Direto no Cerrado,
Goiânia – GO, Brazil.

Vol. 5–2007 91
Fao_75_94.qxp 7-05-2004 15:32 Pagina 92


Vieira, R.C.M.T.; Teixeira Filho, A.R.; Oliveira, A.J. & Lopes, M.R. 2001.
Cadeias Produtivas no Brasil – Análise da Competitividade. Embrapa
Comunicação para Transferência de Tecnologia. Brasília – DF.
Vilela L.; Barcellos, A.O. & Sousa, D.M.G. 2001. Benefícios da Integração entre
Lavoura e Pecuária. Planaltina, DF: Embrapa Cerrados, 2001. 21 p.
(Documentos, 42).
Vilela, L.; Martha Junior, G.B.; Barcellos, A., De O. & Barioni L.G. 2004.
Integração lavoura/pecuária:a sustentabilidade do Cerrado. XXV Congresso
de Milho e Sorgo, 29th Aug – 9th Sept, 2004, Cuiabá-MT Brazil.
Vinne, A.B.V.D. 2001. Integração Agricultura-Pecuária no Sistema Plantio
Direto na Palha. Anais: II Seminário sobre o Sistema Plantio Direto na UFV:
257–164. Universidade Federal de Viçosa – MG, Brazil
Von Schaffhausen, R. 1949. Pigeon pea undersown in maize for improved stubble
grazing. Proceedings of the III International Grassland Conference. São
Paulo State Animal Nutrition Center, Piracicaba-SP, Brazil.
Yokoyama, L.P.; Vianna Filho, A.; Balbino, L.C. & Oliveira, L.P. 1999.
Avaliação, Econômica de técnicas de recuperação de pastagens. Pesquisa
Agropecuária Brasileira 34: 1345–1355.

92 Integrated Crop Management
Fao_75_94.qxp 7-05-2004 15:32 Pagina 93
Fao_75_94.qxp 7-05-2004 15:32 Pagina 94

© FA O 2 0 0 7 Printed in Italy, April 2007
Cop_1083_7mm 10-05-2004 14:29 Pagina 1

ISSN 1020-4555
Tropical crop–livestock systems in conservation agriculture
The Brazilian experience 5 Integrated Crop Management Vol. 5–2007

Tropical crop–livestock systems in conservation agriculture – The Brazilian experience
This publication describes how pasture, fodder and livestock
production have been integrated into conservation agriculture
systems in Brazil’s tropical zones. Vast areas of forest have been
cleared in the tropical areas of Brazil for establishment of pastures
that become unproductive once the native fertility of the soil is
exhausted; this leads to yet more forest clearing.

Integrated crop–livestock zero tillage systems allow for the
sustainable production of high-yielding pasture without further
deforestation; in this system, grazing livestock convert both pastures
Tropical crop–livestock systems
and crop residues into cash. The ability of pasture to build up the
fertility and biological activity of the topsoil is well known. The
in conservation agriculture
economics of the system are discussed and its very positive
ecological effects are described at length.
The Brazilian experience
This publication is geared towards agronomists, advanced farmers,
extension workers and agricultural decision-makers throughout the
tropics and subtropics. It is hoped that the many lessons learned and
technologies developed in the Brazilian tropics can serve, with the
necessary local adaptation, as a starting reference for other tropical
(and subtropical) zones.

ISBN 978-92-5-105692-9 ISSN 1020-4555

9 789251 056929